Geological correlations of East Antarctica with adjoining continents have been puzzling geologists ever since the concept of a Gondwana supercontinent surfaced. Despite the paucity of outcrops because of ice cover, difficulty of access and extreme weather, the past 50 years of Japanese Antarctic Research Expeditions (JARE) has successfully revealed vital elements of the geology of East Antarctica. This volume presents reviews and new research from localities across East Antarctica, especially from Dronning Maud Land to Enderby Land, where the geological record preserves a history that spans the Archaean and Proterozoic. The reviews include extensive bibliographies of results obtained by geologists who participated in the JARE. Comprehensive geological, petrological and geochemical studies, form a platform for future research on the formation and dispersion of Rodinia in the Mesoproterozoic and subsequent assembly of Gondwana in the Neoproterozoic to Early Palaeozoic.
Geodynamic Evolution of East Antarctica: A Key to the East –West Gondwana Connection
The Geological Society of London Books Editorial Committee Chief Editor
BOB PANKHURST (UK) Society Books Editors
JOHN GREGORY (UK) JIM GRIFFITHS (UK) JOHN HOWE (UK) PHIL LEAT (UK) NICK ROBINS (UK) JONATHAN TURNER (UK) Society Books Advisors
MIKE BROWN (USA) ERIC BUFFETAUT (FRANCE ) JONATHAN CRAIG (ITALY ) RETO GIERE´ (GERMANY ) TOM MC CANN (GERMANY ) DOUG STEAD (CANADA ) RANDELL STEPHENSON (NETHERLANDS )
Geological Society books refereeing procedures The Society makes every effort to ensure that the scientific and production quality of its books matches that of its journals. Since 1997, all book proposals have been refereed by specialist reviewers as well as by the Society’s Books Editorial Committee. If the referees identify weaknesses in the proposal, these must be addressed before the proposal is accepted. Once the book is accepted, the Society Book Editors ensure that the volume editors follow strict guidelines on refereeing and quality control. We insist that individual papers can only be accepted after satisfactory review by two independent referees. The questions on the review forms are similar to those for Journal of the Geological Society. The referees’ forms and comments must be available to the Society’s Book Editors on request. Although many of the books result from meetings, the editors are expected to commission papers that were not presented at the meeting to ensure that the book provides a balanced coverage of the subject. Being accepted for presentation at the meeting does not guarantee inclusion in the book. More information about submitting a proposal and producing a book for the Society can be found on its web site: www.geolsoc.org.uk.
It is recommended that reference to all or part of this book should be made in one of the following ways: SATISH -KUMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) 2008. Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308. TOYOSHIMA , T., OSANAI , Y. & NOGI , Y. 2008. Macroscopic geological structures of the Napier and Rayner Complexes, East Antarctica. In: SATISH -KUMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East –West Gondwana Connection. Geological Society, London, Special Publications, 308, 139–146.
GEOLOGICAL SOCIETY SPECIAL PUBLICATION NO. 308
Geodynamic Evolution of East Antarctica: A Key to the East – West Gondwana Connection
EDITED BY
M. SATISH-KUMAR Shizuoka University, Japan
Y. MOTOYOSHI National Institute of Polar Research, Japan
Y. OSANAI Kyushu University, Japan
Y. HIROI Chiba University, Japan and
K. SHIRAISHI National Institute of Polar Research, Japan
2008 Published by The Geological Society London
THE GEOLOGICAL SOCIETY The Geological Society of London (GSL) was founded in 1807. It is the oldest national geological society in the world and the largest in Europe. It was incorporated under Royal Charter in 1825 and is Registered Charity 210161. The Society is the UK national learned and professional society for geology with a worldwide Fellowship (FGS) of over 9000. The Society has the power to confer Chartered status on suitably qualified Fellows, and about 2000 of the Fellowship carry the title (CGeol). Chartered Geologists may also obtain the equivalent European title, European Geologist (EurGeol). One fifth of the Society’s fellowship resides outside the UK. To find out more about the Society, log on to www.geolsoc.org.uk. The Geological Society Publishing House (Bath, UK) produces the Society’s international journals and books, and acts as European distributor for selected publications of the American Association of Petroleum Geologists (AAPG), the Indonesian Petroleum Association (IPA), the Geological Society of America (GSA), the Society for Sedimentary Geology (SEPM) and the Geologists’ Association (GA). Joint marketing agreements ensure that GSL Fellows may purchase these societies’ publications at a discount. The Society’s online bookshop (accessible from www.geolsoc.org.uk) offers secure book purchasing with your credit or debit card. To find out about joining the Society and benefiting from substantial discounts on publications of GSL and other societies worldwide, consult www.geolsoc.org.uk, or contact the Fellowship Department at: The Geological Society, Burlington House, Piccadilly, London W1J 0BG: Tel. þ44 (0)20 7434 9944; Fax þ44 (0)20 7439 8975; E-mail:
[email protected]. For information about the Society’s meetings, consult Events on www.geolsoc.org.uk. To find out more about the Society’s Corporate Affiliates Scheme, write to
[email protected] Published by The Geological Society from: The Geological Society Publishing House, Unit 7, Brassmill Enterprise Centre, Brassmill Lane, Bath BA1 3JN, UK (Orders: Tel. þ44 (0)1225 445046, Fax þ44 (0)1225 442836) Online bookshop: www.geolsoc.org.uk/bookshop The publishers make no representation, express or implied, with regard to the accuracy of the information contained in this book and cannot accept any legal responsibility for any errors or omissions that may be made. # The Geological Society of London 2008. All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No paragraph of this publication may be reproduced, copied or transmitted save with the provisions of the Copyright Licensing Agency, 90 Tottenham Court Road, London W1P 9HE. Users registered with the Copyright Clearance Center, 27 Congress Street, Salem, MA 01970, USA: the item-fee code for this publication is 0305-8719/08/$15.00. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. ISBN 978-1-86239-268-7 Typeset by Techset Composition Ltd., Salisbury, UK Printed by MPG Books Ltd., Bodmin, UK Distributors North America For trade and institutional orders: The Geological Society, c/o AIDC, 82 Winter Sport Lane, Williston, VT 05495, USA Orders: Tel. þ1 800-972-9892 Fax þ1 802-864-7626 E-mail
[email protected] For individual and corporate orders: AAPG Bookstore, PO Box 979, Tulsa, OK 74101-0979, USA Orders: Tel. þ1 918-584-2555 Fax þ1 918-560-2652 E-mail
[email protected] Website http://bookstore.aapg.org India Affiliated East-West Press Private Ltd, Marketing Division, G-1/16 Ansari Road, Darya Ganj, New Delhi 110 002, India Orders: Tel. þ91 11 2327-9113/2326-4180 Fax þ91 11 2326-0538 E-mail
[email protected]
Preface International collaboration to study geophysical phenomena in the polar region dates back to 1882, which was designated as the first International Polar Year (IPY). The second IPY was organized 50 years later in 1932. Thereafter, in view of the progress in observation potentiality and scientific demands, the third IPY was arranged 25 years after the second IPY, and it was renamed as the International Geophysical Year (IGY) and ran during the period from July 1957 to December 1958. Japan commenced its scientific activity in the Antarctic in 1957 as one of the 64 participating nations in IGY, and established ‘Syowa Station’ on the Ongul Islands, Lu¨tzow-Holm Bay in East Antarctica by the 1st Japanese Antarctic Research Expedition (JARE-1) on 29 January 1957. Since then, Japan has been undertaking observation and research projects on various disciplines of natural sciences extending over 50 years in the Antarctic. Geological survey around Syowa Station, a region where no human being had ever set foot, started from the beginning in 1957 by JARE-1. During the past 50 years, a total number of nearly 100 geologists has joined in JARE to conduct geological surveys in the Lu¨tzow-Holm Bay region, Prince Olav Coast, the Yamato Mountains, the Belgica Mountains, the Sør Rondane Mountains, and Enderby Land, in East Antarctica. Despite a paucity of outcrops because of ice cover, difficulty of access and extreme weather, the JARE has successfully revealed vital elements of the geology of East Antarctica, and this led us to the attempt to clarify the origin and evolution of continents and their dynamics in the Earth’s history. Geological correlation of East Antarctica with adjoining continents has been a major topic of discussion among geologists. However, in the early 1990s, JARE succeeded in revolutionizing our understanding of East Antarctic geodynamics by the discovery of a Cambrian mobile belt in Lu¨tzow-Holm Bay. A couple of decades after, through this Special Publication, we attempt to compile reviews and new research from localities across East Antarctica, especially from Dronning Maud Land and Enderby Land. Reviews provide extensive bibliographies of results obtained by geologists who participated in the JARE geological activities. Comprehensive geological, petrological
and geochemical studies will potentially form a platform for future research on the geodynamics of amalgamation of Gondwana in the Neoproterozoic to Early Palaeozoic. In addition, the coincidence of Gondwana amalgamation with major global-scale climatic, environmental and biological changes in Late Neoproterozoic to Early Palaeozoic times implies a close connection between largescale tectonic events and global change, which needs to be confirmed in the future. The success of geological studies of JARE is indebted to the dedicated logistic support of the crews of icebreakers Soya, Fuji and Shirase. The journey is continuing with the commissioning of a new vessel. The last 50 years of dedicated group work of the Japanese Antarctic Research Expeditions in Antarctica is commendable, in this extreme environment of hazardous weather and travel conditions. We hope this Special Publication will enthuse the young generation and be a new starting point for the next 50 years research activity on the Antarctic geosciences.
This publication is a part of the Japanese contribution to IPY 2007–2008. M. SATISH -KUMAR , Y. MOTOYOSHI , Y. OSANAI , Y. HIROI & K. SHIRAISHI
Acknowledgements The editors would like to thank the following scientists around the world who gave generously of their time and expertise in reviewing the manuscripts submitted to this Special Publication. Makoto Arima Sotaro Baba Wilfried Bauer Mike Brown Chris Carson Somnath Dasgupta Prelevic Dejan Christoph Dobmeier Dennis Eberl David Ellis Mike Flowerdew Reinhardt Fuck Geoff Grantham Ed Grew Richard Hanson Joerg Hermann Tomokazu Hokada Julie Hollis Jan-Marteen Huizenga Kiyoshi Ito Masahiro Ishikawa Hideo Ishizuka Joachim Jacobs Simon Johnson Hiroo Kagami Hiroyuki Kagi Masaki Kanao
Ken-ichi Kano Dave Kelsey Tony Kemp Kare Kullerud Axel Liebscher Victor Melezhik Akira Miyake Anand Mohan Tomoaki Morishita Hans Mueller Takashi Nakajima Atsushi Okamoto Masaaki Owada Bob Pankhurst Konstantin Podlesskii H. M. Rajesh V. Ravikant K. Sajeev M. Santosh Rajesh Shrivastava Robert Stern Fabrizio Storti Bob Thomas Nobutaka Tsuchiya Carlos Villaseca Yue Zhao
Finally, we are grateful to Phil Leat, the editor-in-charge of the Special Publication, and Angharad Hills of the Geological Society Publishing House for continuous support and encouragement throughout the editing process.
Contents Preface
vii
Acknowledgements
viii
SATISH -KUMAR , M., HOKADA , T., KAWAKAMI , T. & DUNKLEY , D. J. Geosciences research in East Antarctica (08E–608E): present status and future perspectives
1
SHIRAISHI , K., DUNKLEY , D. J., HOKADA , T., FANNING , C. M., KAGAMI , H. & HAMAMOTO , T. Geochronological constraints on the Late Proterozoic to Cambrian crustal evolution of eastern Dronning Maud Land, East Antarctica: a synthesis of SHRIMP U –Pb age and Nd model age data
21
JACOBS , J., BINGEN , B., THOMAS , R. J., BAUER , W., WINGATE , M. T. D. & FEITIO , P. Early Palaeozoic orogenic collapse and voluminous late-tectonic magmatism in Dronning Maud Land and Mozambique: insights into the partially delaminated orogenic root of the East African– Antarctic Orogen?
69
GRANTHAM , G. H., MACEY , P. H., INGRAM , B. A., ROBERTS , M. P., ARMSTRONG , R. A., HOKADA , T., SHIRAISHI , K., JACKSON , C., BISNATH , A. & MANHICA , V. Terrane correlation between Antarctica, Mozambique and Sri Lanka; comparisons of geochronology, lithology, structure and metamorphism and possible implications for the geology of southern Africa and Antarctica
91
ISHIZUKA , H. An overview of geological studies of JARE in the Napier Complex, Enderby Land, East Antarctica
121
TOYOSHIMA , T., OSANAI , Y. & NOGI , Y. Macroscopic geological structures of the Napier and Rayner Complexes, East Antarctica
139
SATISH -KUMAR , M., MIYAMOTO , T., HERMANN , J., KAGAMI , H., OSANAI , Y. & MOTOYOSHI , Y. Pre-metamorphic carbon, oxygen and strontium isotope signature of high-grade marbles from the Lu¨tzow-Holm Complex, East Antarctica: apparent age constraints of carbonate deposition
147
MIYAMOTO , T., SATISH -KUMAR , M., DUNKLEY , D. J., OSANAI , Y., YOSHIMURA , Y., MOTOYOSHI , Y. & CARSON , C. J. Post-peak (,530 Ma) thermal history of Lu¨tzow-Holm Complex, East Antarctica, based on Rb – Sr and Sm–Nd mineral chronology
165
ISHIKAWA , M., SHINGAI , E. & ARIMA , M. Elastic properties of high-grade metamorphosed igneous rocks from Enderby Land and eastern Dronning Maud Land, Antarctica: evidence for biotite-bearing mafic lower crust
183
SUZUKI , S., ISHIZUKA , H. & KAGAMI , H. Early to middle Proterozoic dykes in the Mt. Riiser-Larsen area of the Napier Complex, East Antarctica: tectonic implications as deduced from geochemical studies
195
SUDA , Y., KAWANO , Y., YAXLEY , G., KORENAGA , H. & HIROI , Y. Magmatic evolution and tectonic setting of metabasites from Lu¨tzow-Holm Complex, East Antarctica
211
OWADA , M., BABA , S., OSANAI , Y. & KAGAMI , H. Geochemistry of post-kinematic mafic dykes from central to eastern Dronning Maud Land, East Antarctica: evidence for a Pan-African suture in Dronning Maud Land
235
HOKADA , T., MOTOYOSHI , Y., SUZUKI , S., ISHIKAWA , M. & ISHIZUKA , H. Geodynamic evolution of Mt. Riiser-Larsen, Napier Complex, East Antarctica, with reference to the UHT mineral associations and their reaction relations
253
CARSON , C. J. & AGUE , J. J. Early Palaeozoic metasomatism of the Archaean Napier Complex, East Antarctica
283
TSUNOGAE , T., SANTOSH , M., DUBESSY , J., OSANAI , Y., OWADA , M., HOKADA , T. & TOYOSHIMA , T. Carbonic fluids in ultrahigh-temperature metamorphism: evidence from Raman spectroscopic study of fluid inclusions in granulites from the Napier Complex, East Antarctica
317
vi
CONTENTS
HIROI , Y., MOTOYOSHI , Y., ISHIKAWA , N., HOKADA , T. & SHIRAISHI , K. Origin of xenocrystic garnet and kyanite in clinopyroxene –hornblende-bearing adakitic meta-tonalites from Cape Hinode, Prince Olav Coast, East Antarctica
333
KAWAKAMI , T., GREW , E. S., MOTOYOSHI , Y., SHEARER , C. K., IKEDA , T., BURGER , P. V. & KUSACHI , I. Kornerupine sensu stricto associated with mafic and ultramafic rocks in the Lu¨tzow-Holm Complex at Akami Point, East Antarctica: what is the source of boron?
351
YOSHIMURA , Y., MOTOYOSHI , Y. & MIYAMOTO , T. Sapphirine þ quartz association in garnet: implication for ultrahigh-temperature metamorphism in Rundva˚gshetta, Lu¨tzow-Holm Complex, East Antarctica
377
GOTO , S. & IKEDA , T. Crystal size distribution of garnet in quartzo-feldspathic gneisses from the Lu¨tzow-Holm Complex at Skallen, East Antarctica
391
BABA , S., OWADA , M. & SHIRAISHI , K. Contrasting metamorphic P–T path between Schirmacher Hills and Mu¨hlig-Hofmannfjella, central Dronning Maud Land, East Antarctica
401
KAWASAKI , T. & OSANAI , Y. Empirical thermometer of TiO2 in quartz for ultrahightemperature granulites of East Antarctica
419
SATO , K., MIYAMOTO , T. & KAWASAKI , T. Fe2þ –Mg partitioning experiments between orthopyroxene and spinel using ultrahigh-temperature granulite from the Napier Complex, East Antarctica
431
Index
449
Geosciences research in East Antarctica (088E – 6088E): present status and future perspectives M. SATISH-KUMAR1, T. HOKADA2, T. KAWAKAMI3 & DANIEL J. DUNKLEY2 1
Institute of Geosciences, Shizuoka University, Oya 836, Suruga-ku, Shizuoka 422-8529, Japan (e-mail:
[email protected]) 2
3
National Institute of Polar Research, Kaga, Itabashi-ku, Tokyo 173-8515, Japan
Department of Geology and Mineralogy, Kyoto University, Kitashirakawa-oiwake-cho, Sakyo-ku, Kyoto 606-8502, Japan Abstract: In both palaeoenvironmental and palaeogeographical studies, Antarctica plays a unique role in our understanding of the history of the Earth. It has maintained a unique geographical position at the South Pole for long periods. As the only unpopulated continent, the absence of political barriers or short-term economic interests has allowed international collaborative science to flourish. Although 98% of its area is covered by ice, the coastal Antarctic region is one of the wellstudied regions in the world. The integrity and success of geological studies lies in the fact that exposed outcrops are well preserved in the low-latitude climate. The continuing programme of the Japanese Antarctic Research Expedition focuses on the geology of East Antarctica, especially in the Dronning Maud Land and Enderby Land regions. Enderby Land preserves some of the oldest Archaean rocks on Earth, and the Mesoproterozoic to Palaeozoic history of Dronning Maud Land is extremely important in understanding the formation and dispersion of Rodinia and subsequent assembly of Gondwana. The geological features in this region have great significance in defining the temporal and spatial extension of orogenic belts formed by the collision of proto-continents. Present understanding of the evolution of East Antarctica in terms of global tectonics allows us to visualize how continents have evolved through time and space, and how far back in time the present-day plate-tectonic regime may have operated. Although several fundamental research problems still need to be resolved, the future direction of geoscience research in Antarctica will focus on how the formation and evolution of continents and supercontinents have affected the Earth’s environment, a question that has been addressed only in recent years.
The formation and evolution of continents has always been an intriguing topic in Earth Science studies. The complexity of continental evolution largely results from the protracted and recurring nature of geological processes that have taken place in the continental lithosphere. Decoding billions of years of complex history recorded in the continental crust is a daunting task. However, geologists have made great progress in understanding the processes involved in continental formation and their evolution through time. The Antarctic continental lithosphere is an important crustal fragment that provides us with an abundance of information on the formation of continents, and the temporal and spatial relationships involved in the assembly and dispersion of supercontinents. The significance of Antarctica lies not only in its unique geographical position, whereby it has gained due importance in palaeoenvironmental studies, but also in its geological stability since incorporation in the supercontinent Gondwana at the beginning of the Phanerozoic Era. This is primarily because the Antarctic lithospheric plate has been surrounded by
mid-ocean ridges since the Mesozoic (Fig. 1 inset), with the exception of the Antarctic Peninsula, which is its only active convergent plate margin with transform faults dividing the Antarctic Plate and Scotia Plate. This means that the Antarctic Plate is currently expanding relative to the surrounding plates. This is a feature that is at variance with most other lithospheric plates, and makes the Antarctic continent exceptionally stable, and isolated from all regional tectonic events during the Mesozoic and Cenozoic. Consequently, the older geological history of East Antarctica can be considered as one of the least overprinted records of crustal evolution in the Earth’s history, preserved in a natural ‘cold storage’. Its geological history records the formation of early Archaean protocontinents, and continues throughout the Proterozoic, until the amalgamation of East and West Gondwana at the beginning of the Palaeozoic. Therefore, the geological record in East Antarctica is an invaluable record of the origin and evolution of continents and supercontinents, and for understanding the secular changes in metamorphic conditions in orogenic belts (Brown 2007).
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 1 –20. DOI: 10.1144/SP308.1 0305-8719/08/$15.00 # The Geological Society of London 2008.
2 M. SATISH-KUMAR ET AL. Fig. 1. Index map of geographical regions and localities in East Antarctica corresponding to the contributions in this Special Publication. Inset shows a topographic map of Antarctica and surrounding oceans. Red indicates topographically elevated places; blue indicates ocean floor. (Data source: Department of Commerce, National Oceanic and Atmospheric Administration, National Geophysical Data Center, 2006, 2-minute Gridded Global Relief Data (ETOPO2v2), http://www.ngdc.noaa.gov/mgg/fliers/ 06mgg01.html).
GEOSCIENCES RESEARCH IN EAST ANTARCTICA
It is beyond the scope of this book to update the reader with the voluminous literature that has been produced in the past few decades on the geology of East Antarctica. However, we make an attempt to integrate the results of some recent studies from the eastern region of the Antarctic continent, where the Japanese Antarctic Research Expedition (JARE) has, over the past 50 years, conducted extensive investigations. We introduce the general geology of the region and summarize what is known to date, and in the process introduce the contributions in this volume. The contributions in the volume are related to the outcrops that are situated between 08E and 608E in Dronning Maud Land and Enderby Land of East Antarctica (Fig. 1). In addition, this paper also attempts to lay down ‘a vision for future’, based on the current status of geological knowledge.
East Antarctica: an integral part of Gondwana The challenge of developing tectonic scenarios for the formation of the ice-covered Antarctic continent is uniquely difficult; no other continent presents such a blank sheet on which geological terranes can be drawn by inference only. Virtually all understanding of the geological architecture is drawn from intensive studies of coastal outcrops and mountain ranges near the continental margins. A full 1808 arc of coastline, encompassing East Antarctica, provides an array of outcrops that almost exclusively share a Precambrian origin. This reflects the intracontinental nature of the East Antarctic coast in the supercontinent Gondwana, after its formation at the end of the Proterozoic. The stability of the continent throughout the Phanerozoic has also led to the concept of an East Antarctic Shield, one of the large areas of cratonized crust on Earth. The ‘shield’ concept also influenced tectonic interpretations of coastal geology before the formation of Gondwana. It was recognized that most localities in East Antarctica are represented by areas of high tectonic activity, dominated by moderate- to high-temperature metamorphic belts, shear zones, and regions of Proterozoic crustal growth, and that Archaean granite– greenstone and metamorphic terranes are mostly restricted to small discrete localities. This led to the development of tectonic models of a ‘cratonized’ East Antarctic Shield with extensive mobile belts, such as the c. 1 Ga Circum-Antarctic Mobile Belt of Yoshida (1992) and the Wegener –Mawson Mobile Belt of Kamenev (1991). These models implied the existence of a coherent Antarctic continent that was amalgamated during the formation of the supercontinent Rodinia at 1.3– 0.9 Ga
3
(Hoffman 1991). However, subsequent years have seen a steady increase in the volume and detail of tectonic and geochronological research from all areas of East Antarctica that has shown a more complex story of the diverse origins of various sectors of the East Antarctic margin, challenging the ‘shield’ paradigm. Late Mesoproterozoic metamorphic terranes located along the Antarctic coast at 308W–358E (the 1100– 1000 Ma Maud Belt), 458E– 708E (the 1000–900 Ma Rayner Complex) and 1008E–1208E (the 1300– 1100 Ma Wilkes Province), were found not only to differ subtly in age, but also to be separated by areas of c. 600–500 Ma moderate- to high-temperature metamorphism and tectonism at Lu¨tzow-Holm Bay (408E) and Prydz Bay (708E; Fitzsimons 2000). Thus, instead of representing a continuous marginal mobile belt, each of the Mesoproterozoic metamorphic terranes could be correlated with discrete mobile belts in South Africa (Namaqua–Natal Belt), India (Eastern Ghats) and South Australia (Albany– Fraser Orogen). Furthermore, it was recognized that a large section of the Maud Belt was reworked by late Neoproterozoic metamorphism and deformation that could be correlated with the extensive East African Orogen, produced by the amalgamation of East and West Gondwana (Jacobs et al. 2003a). Recognition of unrelated pre-Rodinian cratons in East Antarctica was also achieved, with the correlation of the Mawson continent and the Gawler Craton in South Australia (Fanning et al. 1996), and the geochronological characterization of Archaean terranes south of a c. 550 Ma suture zone in the southern Prince Charles Mountains adjacent to Prydz Bay (Boger et al. 2001; Mikhalsky et al. 2001, 2006; Phillips et al. 2006). New studies (e.g. Kelsey et al. 2008) continue to develop the latest paradigm of the assembly of East Antarctica from disparate continental bodies during the late Neoproterozoic formation of Gondwana. In particular, the complexity of crustal development in the sector between 08E and 708E, namely Dronning Maud Land, Enderby Land, Kemp Land and Mac Robertson Land, is the focus of recent and current research. Shiraishi et al. attempt to synthesize a large amount of new geochronological data obtained from eastern Dronning Maud Land, and discuss the variations in age distributions between the lithological units. Magmatic and metamorphic events between 1200 and 500 Ma are identified from zircon geochronology in different regions, providing insights into the formation and assembly of crustal fragments, Neoproterozoic sedimentation, and late Neoproterozoic to Cambrian episodes of metamorphism and magmatism. Shiraishi et al. further consider the geodynamic evolution of eastern Dronning Maud Land on the basis of published and new Nd model ages, which
4
M. SATISH-KUMAR ET AL.
Fig. 2. Neoproterozoic Gondwana showing the cratonic regions and surrounding mobile belts. Simplified after Gray et al. (2008) and modified taking into consideration the Lawyer et al. (1998) tight-fit Gondwana configuration. SL, Sri Lanka; MD, Madagascar; WA, western Australia; EA, eastern Australia; SA, southern Australia; SF, Sa˜o Franscisco; RP, Rio de la Plata.
indicate juvenile extraction of Mesoproterozoic crust in the Sør Rondane Mountains, in contrast to the mixed Archaean and Proterozoic derivation of continental crust in the Lu¨tzow-Holm Complex. In a Gondwanan perspective, these results will shift the attention of geodynamic modelling to eastern Dronning Maud Land, to clarify the complex orogenic processes involved in the amalgamation of East and West Gondwana (Fig. 2). The significance of voluminous plutonic activity in Dronning Maud Land and northern Mozambique is discussed by Jacobs et al., who consider lateral southward extrusion and extensional collapse as the preferred tectonic scenario, potentially as a result of crustal delamination. The correlation of temporal variations with distinct shifts in geochemical affinities of magmatic regimes form the basis of a delamination model in association with orogenic collapse and escape tectonics, as proposed recently by Jacobs & Thomas (2004). The model remains to be tested in relation to Gondwanan geodynamics. An intriguing conundrum of correlation of terranes in Gondwana is examined by Grantham et al. through a detailed comparison of lithological, structural, metamorphic and geochronological data from Mozambique with Sri Lanka and Dronning Maud Land. As a follow-up study to Grantham et al. (2003), a continuing ambitious mapping project of Mozambique has led these workers to
propose a model dominated by nappe tectonics during the 590– 550 Ma period of Gondwana assembly. Tectonic windows in Sri Lanka and Dronning Maud Land are considered as expressions of the c. 600 km displacement of crust from northern Gondwana by mega-thrusts. This concept will be tested by future developments.
Geological outline of East Antarctica (088E – 6088E) Since the proposal of Gondwana and Rodinia reconstruction models by Dalziel (1991) and Hoffman (1991), geoscientists have conceived of the Antarctic continent as a single stable entity between 1000 and 500 Ma. However, recent palaeomagnetic, geological and geochronological studies have recognized several distinct Neoproterozoic orogenic events within the East Antarctic shield, and a new concept has emerged of East Antarctica as a collage of three distinct Mesoproterozoic provenances: the Wilkes (1330–1130 Ma), Maud (1090–1030 Ma) and Rayner (990 –900 Ma) Provinces (Fitzsimons 2000; Meert 2003; and references therein). Moreover, two distinct age populations of 650–550 Ma and 580– 500 Ma have emerged in extensive geochronological datasets from the so-called ‘Pan-African orogeny’ (Fig. 2),
GEOSCIENCES RESEARCH IN EAST ANTARCTICA
involving several discrete crustal blocks in East Antarctica and regions surrounding the East African –Antarctic Orogen (e.g. Jacobs et al. 2003a, b; Meert 2003; Hokada & Motoyoshi 2006). Provinces of Archaean age in East Antarctica are found at ‘Annadagstopane’ in Grunehogna, western Dronning Maud Land; the Napier Complex and Oygarden Islands in Enderby Land; the southern Prince Charles Mountains, Vestfold Hills and
5
Rauer Islands in Mac Robertson Land and Princess Elizabeth Land; the Denman Glacier in Queen Mary Land, and the Mawson Block in Terre Ade´lie (Fig. 3). The Grunehogna terrane is considered as a part of the Archaean Kalahari Craton in southern Africa (Jacobs et al. 1993b), and the Mawson Block has been correlated with the Gawler Craton in southern Australia (Fanning et al. 1996). Altogether, the terranes of East
Fig. 3. Continents that surrounded Antarctica in the Neoproterozoic. Geological entities within East Antarctica are also shown.
6
M. SATISH-KUMAR ET AL.
Antarctica preserve a protracted crustal history from the oldest in the Napier Complex (c. 3850 Ma) to the last episode of post-collision magmatism (c. 450 Ma) in the waning stages of Gondwana amalgamation. Enderby Land lies between longitudes 458E and 608E, and comprises Archaean to early Proterozoic crustal sequences, representing a continental core complex surrounded by Proterozoic– Cambrian mobile belts. To the east of Enderby Land, more than 1000 km of coastal areas in Dronning Maud Land (58E– 458E) comprise late Proterozoic to Cambrian mobile belts (650– 500 Ma) (Fig. 4). This mobile belt has been extrapolated to Prydz Bay (Boger et al. 2001) and as far as western Australia (Fitzsimons 2000; Meert 2003). The only recent equivalent of such an extensive mobile belt is the Cenozoic Alpine– Himalayan orogenic belt. How far are these two orogens comparable, and where do they differ? Although there are countless similarities between the two, the former lacks the expression of low-temperature/ high-pressure metamorphic belts, which would provide clear equivalents to present-day subduction, accretion and collision-related tectonic settings that presumably would predate the amalgamation of continental blocks by the orogen. Within Dronning Maud Land, the lithological contrast between the inland mountain chains of central–eastern Dronning Maud Land (including the Sør Rondane, Belgica and Yamato Mountains) and outcrops along the Soya and Prince Olav
Coasts further east is striking (Fig. 4). The former region is dominated by felsic (granitic, granodioritic and syenitic) orthogneisses and post-tectonic plutons, with lesser mafic lithologies and metasedimentary sequences. In contrast, the latter region (Lu¨tzow-Holm and western Rayner Complexes) consist of voluminous metasedimentary rocks with mafic and calcareous rocks, and relatively little granitic material. It is important that any regional tectonic model accounts for this transition in the makeup of the mobile belt.
The Napier Complex The Napier Complex is one of several Archaean cratonic terranes (Fig. 3) in the East Antarctic continent (e.g. the Grunehogna terrane, the Ruker terrane in the southern Prince Charles Mountains, the Vestfold Hills, the Mawson Block in Terre Ade´lie, the Miller Range and the Shackleton Range), but is unique in being entirely composed of high-temperature granulites. Early Archaean (.3850 Ma) protolith ages have been obtained from tonalitic orthogneisses (Black et al. 1986; Harley & Black 1997; Kelly & Harley 2005), which are the oldest in Antarctica and close to the age of the Earth’s oldest known orthogneiss, the Acasta Gneiss in Canada (4000 Ma, Bowring et al. 1989). These 3850 Ma tonalitic orthogneisses occur at least in two localities (Mt. Sones and Gage Ridge; Harley & Black 1997; Kelly & Harley 2005), and subsequent tonalitic –granodioritic
Fig. 4. Tectonic units in Dronning Maud Land and Enderby Land, East Antarctica, showing salient geological and geochronological features. Dashed lines represent suspect boundaries between the units.
GEOSCIENCES RESEARCH IN EAST ANTARCTICA
magmatism is observed from 3270 Ma (Mt. RiiserLarsen; Hokada et al. 2003) to 2630 Ma (Tonagh Island; Carson et al. 2002). The Napier Complex consists predominantly of tonalitic –granodioritic orthogneiss, but also includes mafic to ultramafic orthogneisses, garnet-bearing peraluminous granitic gneisses, and subordinate quartzo-feldspathic, siliceous and aluminous paragneisses. This lithological diversity indicates a complex and progressive development of the proto-metamorphic terrane, and provides insights into the development of continental crust during the Archaean. However, it is still unclear when and how the various crustal components were brought together, and what types of tectonic processes were functional in the Archaean. What is known is that the crustal components of the Napier Complex shared a common history after 2850 Ma, the timing of the first major regional magmatic–metamorphic event. Following extensive field and laboratory work by geologists of the Australian National Antarctic Research Expedition (ANARE), who established the geological structure and history of this area (see Sheraton et al. 1987, and references therein), JARE has carried out geological fieldwork intermittently throughout the 1980s and 1990s. Ishizuka reviews the voluminous results obtained by various JARE expeditions to the Napier Complex. In addition to the preparation of detailed geological maps (Ishikawa et al. 2000; Osanai et al. 2001), the expeditions focused on the geochemical characterization of different lithological units within the Napier Complex, which represent an admixture of Archaean components with sedimentary, granite– greenstone and tonalite–trondhjemite–granite (TTG) affinities. The review emphasizes important results obtained in subject areas such as the processes of ultrahigh-temperature (UHT) metamorphism, stages of protolith formation and geochemical studies of dykes, to provide constraints on modelling the tectonic evolution of the region. Our basic knowledge of the regional structural features in the Napier and Rayner Complexes is based on the mapping results carried out in the 1960s by the Soviet Antarctic Expedition (SAE; see Kamenev 1972, 1975) and further extensive geological mapping by ANARE until the late 1970s (Sheraton et al. 1987). Toyoshima et al. construct a regional form-line map based on structural data from published maps in the Napier and Rayner Complexes. They identify potential boundaries between different regions, based on the convergence of several structural parameters. The location of major tectonic boundaries is supported by geophysical evidence, as well as detailed field geological data from representative areas. In addition to the geological information from outcrops, the nature and properties of lower crustal
7
materials deduced from geophysical studies are critical in imaging the present-day crust. Ishikawa et al. examine the seismic and elastic properties of lower crustal rocks from Enderby Land and Dronning Maud Land to provide insights into the lower crust of East Antarctica. They suggest a possible predominance of biotite-bearing continental crust. The Napier Complex experienced unusually high temperatures during metamorphism of 900–1100 8C on a regional scale, providing the first recognized instance of UHT metamorphism. Mineral parageneses diagnostic of UHT metamorphism, including sapphirine þ quartz, orthopyroxene þ sillimanite þ quartz and osumilite (Harley & Hensen 1990), have been recognized over a 200 km by 100 km area. The widespread distribution of UHT metamorphism, with estimated peak metamorphic temperatures in excess of 1120–1150 8C at relatively shallow crustal depths of 20–30 km (e.g. Harley & Motoyoshi 2000; Ishizuka et al. 2002; Harley 2004), requires explanation by unusual tectonic models to reasonably explain these crustal conditions. The terrane attracts great interest in how such extreme temperatures can be achieved in the mid- to lower crust, and represents a metamorphic end-member at the opposite extreme from the ultrahigh pressure (UHP) metamorphism found in continent–continent collision zones. UHT metamorphism in the Napier Complex is a phenomenon that never has been found on such a scale anywhere else in the world. Hokada et al. model the thermal and barometric behaviour of the lower continental crust. Based on an extensive analysis of petrological, structural and geochronological data, they estimate the lateral and vertical extent of UHT lithologies, and discuss the difficulties in providing models that can sustain a .1000 8C thermal regime for crustal thicknesses of 4–5 km. It is stressed that an enormous quantity of heat is necessary for achieving this, and that modelling requires an active role for asthenospheric input. Experimental and empirical studies on various chemical systems in metamorphic mineral assemblages are essential in determining the temperatures and pressures prevailing under UHT conditions. The solubility of titanium in quartz under UHT conditions is evaluated with the help of experiments by Kawasaki & Osanai on samples from Bunt Island, from which they develop an empirical geothermometer, and they test it by applying it to selected localities in Enderby Land and Dronning Maud Land. This method should find application in many future studies, as quartz and titaniumbearing minerals such as rutile and ilmenite are common constituents in high-grade metamorphic rocks. Sato et al. examine the partitioning behaviour of Fe2þ and Mg between orthopyroxene and
8
M. SATISH-KUMAR ET AL.
spinel from UHT assemblages. Although the partitioning does not seem to record UHT conditions because of retrograde exchange, the results are reliable indicators of post-peak conditions. Fluid composition is a critical factor that controls metamorphism under UHT conditions without melting the rock. It is essential that the rocks should be anhydrous when the UHT conditions are attained. According to experimental studies (e.g. Johannes & Holtz 1996), even 1 wt% of water in a muscovite granite will lead to complete melting at UHT conditions. Therefore, the precursor rocks should either be essentially dehydrated during prograde metamorphism, or should have been previously anhydrous (by earlier metamorphism). The anhydrous mineral assemblages under UHT conditions were probably sustained by the presence of dry CO2-rich fluid. Characterization of fluids in UHT rocks from the Napier Complex is elegantly carried out by Tsunogae et al., who apply Raman spectroscopy to obtain the precise chemical composition of the fluids that were present during peak metamorphism. The ubiquitous presence of CO2 is demonstrated. Minor amounts of CH4 and N2 are also identified. Intriguingly, carbonate minerals present within the fluid inclusions further provide a unique window into the evolution of fluids during UHT metamorphism. CO2-rich fluid has an important, if not instrumental, role in UHT metamorphism, because it can be an effective heat transfer medium. High-T carbonic fluids from asthenospheric mantle to crust can effectively transfer heat into the crustal rocks, much faster and more easily than thermal conduction or convection. Enderby Land is also characterized by multiple episodes of dyke emplacement (Sheraton et al. 1987). The geochemical and tectonic significance of post-tectonic dykes is studied by Suzuki et al., who identify two distinct generations of dykes at Mt. Riiser-Larsen that exhibit contrasting source characteristics. An earlier 1.9–2.0 Ga generation of dykes is considered to have derived from a mantle wedge source, with possible connections with the continental crust formation of Rayner Complex. The less prominent 1.2 Ga dyke suite has ocean island basalt (OIB) or enriched mid-ocean ridge basalt (E-MORB) affinities. In addition to the emplacement of dykes, Enderby Land is also intruded by early Palaeozoic pegmatites. Carson & Ague evaluate geochemical element mobility associated with the infiltration of aqueous fluids in association with pegmatites, and model the depth of wall-rock metasomatism. They also suggest that the source for pegmatitic melts and aqueous fluids might be the underplating of sedimentary rocks by convergent tectonism between the Rayner Complex and the Napier
Complex, implying an early Palaeozoic timing for the juxtaposition of these terranes in the western part of Enderby Land.
The Rayner Complex The Rayner Complex was originally named for Proterozoic metamorphic lithologies adjacent to the Archaean Napier Complex (Kamenev 1972). It is made up of amphibolite- to granulite-facies orthogneisses and paragneisses, including pelitic, mafic, ultramafic and calcareous layers and boudins. Although the Rayner Complex was originally defined by lithologies south of the Napier Complex in Enderby Land, the main extent of the terrane is recognized to the east in Kemp Land and Mac Robertson Land. In the latter, the terrane is terminated eastwards by the late Neoproterozoic to Cambrian granulites of Prydz Bay, and southwards by metamorphic rocks and granitoids of similar age in the southern Prince Charles Mountains (Boger & Wilson 2005). The Rayner Complex involves the 990– 900 Ma granulite-grade reworking of supracrustal lithologies, deposited on a basement that is mostly Palaeoproterozoic in the eastern section, with an Archaean component closer to the Napier Complex (Kelly et al. 2002; Halpin et al. 2005). Extensive intrusions of charnockite were emplaced during metamorphism along the eastern margin of the Rayner Complex (Young & Black 1991). The grade and timing of metamorphism and charnockitic magmatism, along with the nature of protolithic crust, are shared with the Eastern Ghats of the Indian peninsula, and the two terranes are now regarded as having been a single tectonic entity attached to the cratonic core of India before the Neoproterozoic (e.g. Dobmeier & Raith 2003). Metamorphic reworking of the eastern Rayner –Eastern Ghats terrane in the late Neoproterozoic is limited to its margins (east and south in the Rayner Complex, north in the Eastern Ghats; Mezger & Cosca 1999). The geological evolution of the western part of the Rayner Complex is more problematic. A predominance of early Cambrian ages (Shiraishi et al. 1997; Motoyoshi et al. 2006) suggests that this area was reworked simultaneously with the metamorphism of the adjacent Lu¨tzow-Holm Complex. Metamorphic conditions, involving isothermal decompression after UHT peak metamorphism at Forefinger Point, are similar to those at Rundva˚gshetta, at the opposite end of the Lu¨tzowHolm Complex. However, major differences between the Lu¨tzow-Holm and western Rayner complexes include the observation of prograde kyanite inclusions in garnet, and the eastwarddecreasing grade of metamorphism in the former terrane. In addition, 800–700 Ma ages obtained in
GEOSCIENCES RESEARCH IN EAST ANTARCTICA
the western Rayner Complex (Asami et al. 1997, 2005; Shiraishi et al. 1997) have not been found in the Lu¨tzow-Holm Complex. Regardless, the extensive Cambrian reworking leads us to redefine this region as the ‘Western Rayner Complex’, in contrast to the main body of the Rayner Complex and the Lu¨tzow-Holm Complex (Fig. 4). The geological significance of the Western Rayner Complex is a subject of continuing and future research, and further field and analytical studies are required to understand this complicated section of East Antarctica.
The Lu¨tzow-Holm Complex The Lu¨tzow-Holm Complex, located in eastern Dronning Maud Land (Fig. 4), is a late Neoproterozoic orogenic belt bounded by the late Mesoproterozoic Rayner Complex to the east and by the late Neoproterozoic to early Palaeozoic Yamato– Belgica Complex to the west (Shiraishi et al. 1992, 1994, 2003). It is a significant area for the investigation of the final collision between East and West Gondwana, because the Lu¨tzow-Holm Complex is considered to be a southern extension of the suture between them (e.g. Shiraishi et al. 1994; Fitzsimons 2000). The geology of this complex has been reviewed in several earlier studies (Hiroi et al. 1983, 1986, 1987, 1991; Shiraishi et al. 1994, 2003). The Lu¨tzow-Holm Complex is composed of high-grade metamorphic rocks, including pelitic– psammitic gneisses, mafic to intermediate basic gneisses, subordinate lenses of ultramafic gneiss, marbles and calc-silicate rocks. Felsic pegmatitic dykes discordantly intrude the metamorphic rocks. Ultramafic lenses that were probably derived from oceanic crust are distributed across the central and southwestern part of the complex (Hiroi et al. 1986). Hiroi et al. (1991) postulated that the ultramafic lenses represent dismembered fragments of an ophiolite complex derived from the missing oceanic crust between older continents, now represented by the Yamato– Belgica and Rayner Complexes. The detailed structural evolution of the Lu¨tzow-Holm Complex has not yet been fully understood, although some parts of the complex have been structurally described in several studies (e.g. Kizaki 1962, 1964; Ishikawa 1976; Yoshida 1977, 1978; Matsumoto et al. 1979, 1982; Ishikawa et al. 1994; Motoyoshi & Ishikawa 1997; Ikeda & Kawakami 2004; Kawakami & Ikeda 2004a, b; Michibayashi et al. 2004; Osanai et al. 2004; Okamoto & Michibayashi 2005). The metamorphic grade of the complex progressively increases from upper amphibolite facies on the Prince Olav Coast to granulite facies in Lu¨tzow-Holm Bay (Hiroi et al. 1991), with a
9
‘thermal axis’ of maximum peak temperature estimated to lie at the southern end of Lu¨tzow-Holm Bay, near Rundva˚gshetta (Motoyoshi 1986). Several lines of evidence suggest that the Lu¨tzow-Holm Complex has experienced a typical ‘clockwise’ P –T path. These include prograde kyanite and staurolite as relict inclusions in garnet or plagioclase (Hiroi et al. 1983; Motoyoshi 1986; Kawakami & Motoyoshi 2004; Satish-Kumar et al. 2006b), and reaction textures in ultramafic rocks (Hiroi et al. 1986) are also significant. It has been observed that paragneisses from the Prince Olav Coast experienced the reaction staurolite ¼ garnet þ aluminosilicate þ spinel þ H2O within the sillimanite stability field, whereas those from Lu¨tzow-Holm Bay experienced the reaction in the kyanite stability field (Hiroi et al. 1983, 1987). This petrographical evidence is peculiar among high-grade metamorphic terranes in East Antarctica, as no obvious prograde P –T paths have been reported except for the Lu¨tzow-Holm Complex (Harley & Hensen 1990). UHT peak metamorphic conditions of about 1000 8C and 11 kbar, and subsequent isothermal decompression have been reported from Rundva˚gshetta (Kawasaki et al. 1993; Ishikawa et al. 1994; Motoyoshi & Ishikawa 1997). Yoshimura et al. present further petrological evidence for UHT metamorphism at Rundva˚gshetta (Fig. 1). The coexistence of sapphirine and quartz within garnet porphyroblasts, high Al contents of orthopyroxene and temperature estimates based on ternary feldspar thermometry suggest that the rocks in this region were metamorphosed above temperatures of 1000 8C. In the neighbouring Skallen region (Fig. 1), Goto & Ikeda present crystal size distributions (CSDs) of garnet in quartzo-feldspathic gneisses metamorphosed at above 800 8C. They attempt to provide reasons for the differences in garnet nucleation and growth between layers. Based on the crystal size distribution of garnet they predict less predominance of Ostwald ripening, even at granulite-facies conditions, in the absence of fluids. The timing of the peak regional metamorphism has been estimated by sensitive high-resolution ion microprobe (SHRIMP) U –Pb zircon dating at between 521 + 9 and 553 + 6 Ma (Shiraishi et al. 1992, 1994, 2003). Zircon from syn-deformational leucosome has a U –Pb age of 517 + 9 Ma, which is interpreted as a melt crystallization age (Fraser et al. 2000). Fraser et al. (2000) suggested from combined SHRIMP zircon analyses and Ar –Ar hornblende and biotite chronology that post-peak decompression and subsequent cooling to c. 300–350 8C took place within a time interval of c. 520–500 Ma. A summary of recent dating results, has been given by Nishi et al. (2002) and references therein. Recently, however, in situ
10
M. SATISH-KUMAR ET AL.
monazite chemical Th –U –total Pb isochron method (CHIME) dating and zircon SHRIMP dating combined with the microstructural observation of monazite and zircon by Hokada & Motoyoshi (2006) yield ages of 650 –580 Ma and 550– 520 Ma for monazite in garnet-bearing felsic gneisses from the Skallen region. Based on the medium to heavy REE (MREE –HREE)-enriched nature of 650 –580 Ma monazite, Hokada & Motoyoshi interpreted the older ages as monazite growth under prograde, garnet-absent conditions, whereas the 550– 520 Ma age group represents monazite grown at peak metamorphism in the presence of garnet. Dunkley (2007), reporting a similar spread of ages from 600 to 500 Ma, interpreted the age range as also reflecting the progressive growth of zircon at various stages during a single clockwise P–T history of the complex. These contrasting interpretations will be tested in the near future by petrological and microstructural studies, to find out whether the unexpectedly long duration of a single metamorphism in the Lu¨tzowHolm Complex is feasible (Dunkley 2007). Miyamoto et al. review the chronology of events after peak metamorphism, and present new Sm– Nd and Rb – Sr ages for key metamorphic rocks in the southwestern Lu¨tzow-Holm Complex. Two possible explanations are put forward for postmetamorphic thermal perturbations in the region, involving either cooling and uplift of the terrane, or reheating by magmatic and associated metasomatic activity. Proterozoic and Archaean detrital cores of zircon grains from Rundva˚gshetta and West Ongul Island (Shiraishi et al. 1994; Fraser 1997) demonstrate ancient provenance in the metasediments of the Lu¨tzow-Holm Complex. Satish-Kumar et al. focus on isotopic compositions and geochemical characteristics of high-grade marbles from the Lu¨tzow-Holm Complex. From earlier studies the inferred depositional ages of sedimentary protoliths in the Mozambique Ocean that separated East and West Gondwana is some time between c. 630 Ma (the earliest metamorphic age reported by Hokada & Motoyoshi 2006) and the youngest Sm–Nd model age of c. 850 Ma (Shiraishi et al.). Carbon, oxygen and strontium isotopic compositions indicate that most metacarbonate rocks were altered by multiple episodes of fluid activity, related to pre-peak, peak and post-peak metamorphic events. By applying multiple geochemical criteria, nearpristine sedimentary signatures were identified in some layers, which when compared with the nonmetamorphic chemostratigraphic curves suggest a depositional age between 830 and 730 Ma. Along the Prince Olav Coast, Cape Hinode (Fig. 1) is an exceptional outcrop where the late Neoproterozoic ages are completely absent and
only a c. 1000 Ma age has been reported (Shiraishi et al. 1994, 2003). Grenvillian ages have been reported from three other localities from the Lu¨tzow-Holm Complex, including Skavsnes (Fraser 1997), Telen and Innhovde (Shiraishi et al. 2003). All of these represent inherited cores of zircon with magmatic zoning, and no c. 1000 Ma metamorphic overgrowths have been found. Therefore, Shiraishi et al. (2003) interpreted the c. 1000 Ma age as representing localized igneous activity. Hiroi et al. (2006) have suggested, on the basis of U – Th–Pb ages reported by Shiraishi et al. (1994, 2003) and Motoyoshi et al. (2004), that the gneisses of Cape Hinode are exotic to other parts of the Prince Olav Coast. Xenocrystic garnet and kyanite in adakitic trondhjemites and tonalities from Cape Hinode are treated by Hiroi et al. as phases that were entrained in Mesoproterozoic tonalitic magmas. Kyanite is a stable matrix phase in Cape Hinode metapelites, contrary to the mode of occurrence of kyanite as relic inclusions within garnet in other parts of the Lu¨tzow-Holm Complex. The lack of 600–500 Ma ages from Cape Hinode also supports the notion of an allochthonous block emplaced in the waning stages of amalgamation of East and West Gondwana. The major age population of 1080– 1000 Ma reported from Cape Hinode is comparable with that of the Maud Province to the west, rather than with the 990–900 Ma ages of the closer Rayner Complex. Continuation from Cape Hinode to the Vijayan Complex of Sri Lanka is possible, with extensions to Mozambique and the Natal Belt (Hiroi et al. 2006). Alternatively, Cape Hinode may represent an isolated block, as implied by gravity and geomagnetic data (Nogi et al. 2006). Pre- to syn-metamorphic granitic rocks are characterized by the irregular shape of the intrusive boundaries, intergradational contacts and intense deformation, and only one of them from Breidva˚gnipa has been dated at 576 Ma by the Rb –Sr whole-rock isochron method (Shimura et al. 1998). Post-tectonic granitic dykes intrude across gneissic fabrics throughout the Lu¨tzowHolm Complex. These granites have been dated by the Rb–Sr whole-rock isochron method as younger than 500 Ma (e.g. Nishi et al. 2002; Ajishi et al. 2004). Suda et al. (2006) carried out a geochemical study of metabasites (mostly garnet-absent) in the Lu¨tzow-Holm Complex. Suda et al. further studied the geochemical and isotopic composition of metamorphosed ultramafic and mafic rocks, and distinguish those of the eastern part of the Lu¨tzow-Holm Complex as derived from immature continental crust during the Mesoproterozoic, from those in the western part as products derived from a matured crust. These results further establish the changing tectonic
GEOSCIENCES RESEARCH IN EAST ANTARCTICA
environment in eastern Dronning Maud Land during the Neoproterozoic. Because of the systematic and gradual southwestward increase of metamorphic grade from upper amphibolite facies to UHT conditions, the Lu¨tzow-Holm Complex provides a good example for study of the behaviour of melts, fluids and accessory minerals under these conditions. Satish-Kumar et al. (2006a) studied scapolite boudins from Skallen and presented detailed petrographical and geochemical evidence for changing fluid composition from scapolite phase equilibria. Kawakami et al. (2006) reported the mode of occurrence of sulphide minerals throughout the Lu¨tzow-Holm Complex and found that sulphide inclusions are completely different in composition and species from those in the rock matrix, retaining information from peak metamorphism. Inclusion sulphides were mostly restitic in composition, suggesting the loss of sulphide melt from the rocks of the Lu¨tzowHolm Complex during anatexis. Kawakami et al. characterize the occurrence of kornerupine in mafic and ultramafic rocks from Akarui Point. They propose possible sources for boron through aqueous fluids derived from sediments or hydrothermal alteration of protoliths by seawater.
The Yamato – Belgica Complex The Yamato –Belgica Complex is also thought of as a late Neoproterozoic to Cambrian orogenic terrane between the Lu¨tzow-Holm Complex and the Sør Rondane Mountains (Fig. 4). It consists of two inland mountain ranges, the Yamato and Belgica Mountains. The area is characterized by widespread granite and syenite intrusions with minor amphibolite-facies metamorphic rocks of quartzo-feldspathic and intermediate composition (Shiraishi et al. 1994). Rare granulite-facies rocks with peak metamorphic conditions of 700 –750 8C and ,5 kbar are found, but the relationship between amphibolite-facies and granulite-facies rocks is uncertain. Age constraints for this area are mainly from zircon SHRIMP data by Shiraishi et al. (1994, 2003) that range from 1000 to 500 Ma, with the exception of one spot yielding an age of c. 2500 Ma. Quartz monzonite and granitic gneiss from the Yamato Mountains yielded an age of 535 Ma, which is interpreted as the timing of amphibolite-facies metamorphism and magmatism. These events followed the widespread syenite magmatism of the area, but the actual timing is not well constrained. Although there are not enough data available to establish the Proterozoic– Cambrian history of this area, the lack of essential .1000 Ma ages suggests juvenile crustal formation in the late Mesoproterozoic, similar to the Sør
11
Rondane Mountains and Central Dronning Maud Land to the west, and in marked contrast to the Lu¨tzow-Holm Complex.
The Sør Rondane Mountains Outcrops in the Sør Rondane Mountains are dominated by Mesoproterozoic crustal lithologies that vary from predominantly arc-related material to continental materials from north to south (Shiraishi et al. 1991; Grew et al. 1992; Osanai et al. 1992). A semi-ductile shear zone divides the region into a northeastern granulite-facies terrane and a southwestern amphibolite-facies terrane. Recently, Asami et al. (2007) estimated peak granulite-facies metamorphism at temperatures of 860– 895 8C and pressures of around 12 kbar for the NE terrane. Furthermore, they found evidence for retrograde metamorphism under amphibolite-facies conditions. Extensive geochronological results presented by Shiraishi et al. suggest that crustal formation in the Sør Rondane Mountains occurred in the late Mesoproterozoic, and that the NE and SW terranes were juxtaposed around c. 570 Ma under amphibolite-grade metamorphic conditions, subsequent to higher temperature metamorphism at c. 600 Ma that affected only the NW terrane. An exact picture of late Neoproterozoic to Cambrian terrane amalgamation and tectonic evolution of the Sør Rondane Mountains requires further field studies, which are being conducted by JARE between 2007 and 2010.
Central Dronning Maud Land High-grade metamorphic rocks intruded by voluminous igneous bodies form coastal and inland mountainous outcrops in central Dronning Maud Land (CDML), from 28 to 148E (Dallmann et al. 1990). Metamorphic rocks in this region comprise banded gneisses and migmatites, whereas igneous rocks are mainly of charnockitic, syenitic and granitic composition (Ohta 1999). Two tectonothermal events have been distinguished in the region, at c. 1100 Ma and between 560 and 490 Ma (Jacobs et al. 1998, 2003a; Paulsson & Austrheim 2003). The younger event is generally considered as part of the East African–Antarctic Orogeny and involves an early collisional event at c. 560 Ma followed by large-scale extension associated with voluminous granitic magmatism. A variety of rock types are found in the CDML, including pelitic granulites, garnet-, biotite- and/or hornblendebearing gneisses, charnockites, mafic granulites and calc-silicate rocks. An early Grenvillian age (c. 1150 Ma) for granulite-facies metamorphism, followed by amphibolite-facies metamorphism at c. 560 Ma, is ascribed to these rocks. In addition,
12
M. SATISH-KUMAR ET AL.
c. 630 Ma ages have been obtained from the coastal outcrop at Schirmacher Oasis, suggesting a different evolution of this area in the late Neoproterozoic compared with that of the inland mountains. A recent study by Bisnath et al. (2006) proposed a two-stage collision model, involving an initial arc– continent collision followed by continent– continent collision. Baba et al. compare the metamorphic evolution of Schirmacher Hills with that of Mu¨hligHofmanfjella and find that, although there is no clear difference in peak P–T conditions, the retrograde P–T paths contrast between these two regions. They suggest that the Schirmacher Hills could be part of SE Africa, whereas the inland mountain regions were part of crust formed during the final amalgamation of East Gondwana. Owada et al. consider the geochemical characteristics of post-tectonic mafic dykes and find that the parental magma was derived from a metasomatized mantle source. Based on a detailed evaluation of Sr and Nd isotope systematics of CDML and the Sør Rondane Mountains, they suggest the possibility of a suture zone of East and West Gondwana transition between these two regions.
Emerging thoughts and future perspectives Geophysical studies The continuing compilation of aeromagnetic, marine and satellite-based surveys by the Antarctic Digital Magnetic Anomaly Project (ADMAP) provides the best picture of the internal architecture of East Antarctica, with the latest versions of the East Antarctic magnetic anomaly map and the Antarctic Digital Magnetic Anomaly Map published by Golynsky (2007) and von Frese et al. (2007), respectively (http://www.geology.ohio-state.edu/ admap/). Geodynamic models of the assembly of various terranes between and during cycles of supercontinent formation need to take into account the regional-scale structural information that magnetic anomaly maps provide. A belt of high magnetic anomalies that curves around the coastal Grunehogna craton in western Dronning Maud Land is correlated with the c. 1.1 Ga Namaqua–Natal mobile belt in South Africa, which shows a similar pattern of anomalies (Golynsky & Jacobs 2001). In contrast, a broad area of low magnetic signature extends across central and eastern Dronning Maud, which corresponds well to the interpretation of this region as Mesoproterzoic felsic crust incorporated into the broad East African– Antarctic Orogen (Jacobs et al.; Shiraishi et al.). This geomagnetic domain
has an abrupt north –south trending termination against a region of positive magnetic anomalies, just east of the Yamato Mountains, that corresponds exactly to the terrane boundary between the Sør Rondane Mountains and Yamato– Belgica Complex and the Lu¨tzow-Holm Complex inferred by Shiraishi et al. However, in other key areas of Late Neoproterozoic geological activity, especially around Prydz Bay and Lu¨tzow-Holm Bay, there is a significant discrepancy between the latest tectonic models made on the basis of surface geology (field geology, petrography and geochronology) and geophysics (aeromagnetic mapping). Golynsky et al. (2002) and Golynsky (2007) suggested that the presence of intense east –west linear anomalies, which extend across the Lambert Graben from the northern Prince Charles Mountains to Prydz Bay, and from the southern Prince Charles Mountains to the Grove Mountains, implies a tectonic association of these areas that predates late Neoproterozoic activity in the region. These features were associated by Golynsky (2007) with paired east – west-trending belts of negative and positive anomalies that extend from Prydz Bay to Lu¨tzow-Holm Bay, where the boundaries of these belts rotate into a trend perpendicular to the Prince Olav Coast. These belts are interpreted as late Mesoproterozoic terranes, corresponding to the Rayner Complex, that suture together the Archaean terranes of the Napier Complex and the Ruker terrane. The model implies that late Neoproterozic metamorphism and magmatism observed in Prydz Bay and the southern Prince Charles Mountains is unrelated to that found in Lu¨tzow-Holm Bay. Counter to tectonic models by Boger et al. (2001) and Phillips et al. (2006) that involve the collision of an Indo-Antarctic continent with inner Antarctica during the formation of Gondwana, Golynsky (2007) attributed all late Neoproterozoic activity to within-plate processes, similar to the concept proposed for the Grenvillian circumAntarctic mobile belt (Yoshida 1992, 2007). However, such a interpretation neglects evidence of the heterogeneous nature of crust modified by late Proterozoic metamorphism in the Lu¨tzowHolm Complex, as indicated by a diversity of protolith and crustal model ages (Shiraishi et al.; Suda et al.), and continental to oceanic geochemical signatures (Satish-Kumar et al.; Suda et al.; Hiroi et al.). In the future, integration of geomagnetic data with surface petrology and geochronology should resolve these issues. Another technique to obtain basement geological data inland is ice core drilling that continues to reach basement rocks. There is a two-fold benefit involved in continental ice core drilling, as both palaeoenvironmental and palaeocontinental problems can be solved in a single project. Ice
GEOSCIENCES RESEARCH IN EAST ANTARCTICA
core drilling projects at Vostok and Dome Fuji have returned promising results and technical know-how on pursuing drilling in subzero conditions. In fact, the Vostok drilling has succeeded in collecting sediment from the bottom of the ice sheet, and preliminary SHRIMP dating of zircon and monazite yielded a range of ages between 1.8 and 0.6 Ga, similar to those seen in coastal mobile belts and a further indication of the pervasive involvement of Mesoproterozoic and Neoproterozoic geological activity in the formation of East Antarctica (Rodionov et al. 2006). To solve the problems of palaeocontinental uncertainties relating to obscurity of the inland Antarctic continent it will be necessary to gather information from inland regions.
Field-based studies Because of the low-latitude climate, lack of rainfall, and absence of vegetation (excepting mosses and lichen), outcrops in Antarctica provide high-quality field information for geological studies. Mechanical
13
weathering by the action of wind and glacial abrasion, and limited chemical or hydrothermal alteration, results in the exposure of fresh outcrops that are perfectly suited for multidisciplinary geological studies (Fig. 5). We have identified the following key localities in eastern Dronning Maud Land, which need further attention to improve our understanding of the Archaean to early Palaeozoic evolution of East Antarctica. Enderby Land is a potentially important area for studies not only for clarifying the tectonism in the Archaean but also for understanding lower crustal processes. This region can enlighten us further about: (1) the formation of continental crust in the Archaean; (2) the causes and consequences of unusually high-temperature (.1000–1150 8C) metamorphism in the Napier Complex; (3) Proterozoic suturing between the Archaean cratons of India (e.g. Dharwar –Napier) and Archaean terranes in the southern Prince Charles Mountains of Antarctica; (4) subcontinental mantle dynamics,
Fig. 5. Illustrative outcrops in East Antarctica, and their potential for future research. (a) Field photograph showing the regional distribution of UHT metamorphic rocks in the Napier Complex at Tonagh Island (JARE-38). The Napier Complex is a key area in understanding the crustal evolution in the Archaean. (b) Metacarbonate and paragneiss sequences at Skallevikshalsen (JARE-46), with potential for understanding the depositional environment of sediments between East and West Gondwana. (c) Partial melting and melt segregation as seen in the paragneisses at Skallevikshalsen (JARE-44); this is a topic of prime importance for understanding the generation, segregation and movement of melts in middle to deep continental crust. (d) The inland nunataks of the Sør Rondane Mountains (JARE-49). Geological evidence from these nunataks may clarify the history of amalgamation of East and West Gondwana.
14
M. SATISH-KUMAR ET AL.
as revealed by Proterozoic dyke swarms; (5) the amalgamation of the Napier and Rayner Complexes with other terranes in the formation of Gondwana. Moving west, the Lu¨tzow-Holm Complex has well-preserved regional amphibolite to UHT metamorphic zones with classic clockwise P –T trajectories. The problems that remain to be solved include: (1) the unravelling of the earlier peak granulite to UHT metamorphism and later extensive rehydration; (2) the significance of dual 600– 550 and 550– 500 Ma events in a regional context of Gondwana amalgamation; (3) the provenance and tectonic setting of volcano-sedimentary sequences and basement lithologies. The Yamato and Belgica Mountains are constituted mostly of felsic orthogneisses and syntectonic plutonic rocks. There are fewer suitable lithologies for the detailed characterization of the metamorphic P–T evolution for comparison with the neighboring Lu¨tzow-Holm Complex and the Sør Rondane Mountains. However, preliminary information regarding the geochemical features of plutonic rocks needs to be further developed to determine if the Mesoproterozoic juvenile crust identified in central Dronning Maud Land and the Sør Rondane Mountains extends to this area. Further west, the Sør Rondane Mountains is an important area from a regional geological point of view. This area is seen as critical in finding solutions to longstanding problems on the suturing of East and West Gondwana. The variety of ages recorded in this region may be critical in distinguishing the 600– 550 and 550 –500 Ma conjugate tectonic belts and the order in which crustal fragments amalgamated to form Gondwana.
Understanding geological extremes The geology of East Antarctica not only has provided a regional framework for supercontinent correlation studies but also has been critical in understanding some of the most extreme geological phenomena in crustal regimes. One such extreme is UHT metamorphism. It is increasingly accepted that UHT conditions exist in the continental crust; however, it is still a challenge to understand the factors that control such unusual thermal regimes. In this perspective, more accurate physical parameters and much tighter temporal constraints of such extreme conditions need to be determined. It is also essential to understand the total heat budget, the quantity of heat added to a certain initial or steady-state condition, and whether other factors such as fluids played a role. Precise determination of physical conditions under extreme crustal metamorphism is essential in modelling crustal evolution. It is a challenge to duplicate these conditions in the laboratory,
although recent developments in experimental petrology can achieve this, except for the time factor. Microstructures in minerals, especially exsolution textures, are now recognized as powerful tools for recovering high-pressure and -temperature conditions prior to cooling and exhumation. Typical examples are the recovery of pigeonite compositions from orthopyroxene with Ca-clinopyroxene lamellae (e.g. Harley 1987; Ishizuka et al. 2002), recovery of single-phase compositions from ternary mesoperthitic feldspar (Hokada 2001) and Ti exsolution in quartz or in garnet (Kawasaki & Osanai). However, this technique needs caution in selecting suitable compositional ranges to recover such information and thermodynamic models to be applied for temperature estimation (e.g. Hokada & Suzuki 2006). The formation and preservation of UHT rocks in the crust is essentially controlled by the fluid regime during prograde metamorphism. Dehydration of rocks prior to partial melting is essential to restrict the melt fraction to a critical melting proportion, as larger proportions of melt can destroy the solid rock structure. In other words, UHT metamorphism should be observed only in rocks that are more or less in restitic nature, and potentially anhydrous UHT rocks may be widely distributed in the deepest continental crust worldwide. Composition of fluid also strongly controls the melt fraction. CO2-rich fluid flow from deeper (and hotter) crust transfers heat to shallower crust more effectively than conduction or convection. In addition, we also need to pay attention to the different cooling and uplifting processes that result in the exposure of extremely metamorphosed rocks without completely destroying the original parageneses. UHT metamorphism with subsequent isothermal decompression can be readily achieved by crustal uplift, and internal radioactive heat production in thickened crustal is a potential source of heat. In contrast, UHT metamorphism with isobaric cooling is problematic; that is a scenario that may be achieved when the heat source is local and magmatic (e.g. Bamble terrane in Sveconorwegian or Wilson Lake in Canada, where UHT metamorphic zones are developed around anorthosite bodies). Therefore, our fundamental understanding of extreme crustal processes remains primitive. The Napier Complex in East Antarctica is perfectly suited to understand the occurrence and importance of geological extremes.
Nanoscience and supercontinents: recent technological realms The past two decades have seen wide application of electron microprobe and ion microprobe techniques to investigate the chemical and isotopic
GEOSCIENCES RESEARCH IN EAST ANTARCTICA
composition of minerals, especially the accessory phases, on a micrometre scale. Accessory phase behaviour with regard to trace element geochemistry (including REE, P, Zr, Ti, U and Th) is a major topic in the microanalytical world and has potential in resolving many problems relating to the evolution of continents. In recent years, barriers have been broken in linking the isotopic record with petrology in complex and multiply metamorphosed and deformed terranes (Rubatto 2002; Mu¨ller 2003; Vance et al. 2003). Within our reach is a new phase in accessory mineral research that will unravel complicated metamorphic and tectonic histories. Submillimetre-scale techniques are required to distinguish events in the multiply reactivated mobile belts of East Antarctica. Effective strategies include: (1) U –Th –Pb dating on a sub-grain, micrometre scale of zircon, monazite, apatite, titanite, rutile, perrierite and other accessory minerals by ion microprobe, electron microprobe, and laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS); (2) tying in the microstructural context of accessory phase growth with geochronology and chemistry, using highresolution secondary electron, back-scattered electron, and cathodoluminescence imaging under a scanning electron microscope; (3) experimental and empirical approaches to understanding the stability and chemical behaviour of accessory phases during deformation, metamorphism and partial melting (e.g. Harrison & Watson 1983; Watson & Harrison 1983); (4) integration of metamorphic and magmatic ages obtained by microbeam techniques with Lu –Hf isotope model ages for understanding the crustal extraction history (e.g. Kemp et al. 2006); (5) understanding Precambrian crustal evolution from non-radiogenic isotopes such as oxygen (e.g. Cavosie et al. 2005). Application of these advanced analytical techniques in East Antarctica will help in formulating reasonable geodynamic models of pre-Gondwanan supercontinent evolution.
Supercontinent cycle, global tectonics and Earth’s environment The Neoproterozoic to early Cambrian period was a time of extensive global tectonic activity that culminated in the amalgamation of the supercontinent Gondwana. This time span is also well known for phenomenal changes in climatic conditions that predated the Cambrian explosion in biodiversity. However, extreme climate change models invoking a ‘Snowball Earth’ (Hoffman et al. 1998) or drastic changes in Earth’s obliquity (Evans 2000) do not seem to satisfactorily explain the complex scenario (Meert 2007). Several lines of evidence have
15
started to appear for the role of global tectonism in creating an environment conducive to biological activity (e.g. Maruyama & Santosh 2008; Meert & Liebermann 2008; Stern 2008). Undisputedly, a key factor that controls global environment is CO2 concentration in the atmosphere. Although volcanogenic-CO2 input to the atmosphere seems to be a potential source of sudden large-scale climatic variations, other sources such as CO2 transfer through orogenesis and oxidation of an earlier biosphere cannot be neglected. Variations in atmospheric CO2 in the past are clearly recorded by carbon isotope excursions in carbonate sediments that record conditions in palaeo-oceans. Examples of Neoproterozoic carbon isotopic excursions combined with geological evidence convincingly indicate two major and several minor glaciation events (e.g. Halverson et al. 2005). It still remains unclear how much the closure of oceans between the continents has a bearing on global climate change. Furthermore, it is perceived that the spatial extent of Neoproterozoic orogenic belts retained in the present day continental crust must have been a few orders of magnitude larger than what we see in the Cenozoic Alpine –Himalayan Orogeny. The impact of Neoproterozoic amalgamation of the Gondwana supercontinent on the global environment is yet to be clarified and information from East Antarctica is crucial in solving this problem.
Concluding remarks Studies on Antarctica have considerably refined our knowledge on the geodynamic evolution of continental crust. We envisage Antarctica as a model in Earth Science studies, in the advancement of science and for the peaceful living of mankind. However, because of its remoteness and extreme weather, Antarctica is still a difficult place to carry out geological fieldwork. As discussed above, many fundamental problems remain unsolved. However, progress through international collaboration can efficiently tackle this handicap. The future of Antarctic geoscience research seems bright through collective effort from different countries, and will be a driving force for the advance of our understanding of the history of the Earth. Enormous progress has been achieved in the past 50 years of geological research in East Antarctica. However, technology has overwhelmingly overtaken the pace of basic scientific research. The incongruity between basic scientific research and the momentum with which information is provided by the latest technology is challenging the whole world of science itself. Geoscience research is no exception to this trend. Being part of the natural
16
M. SATISH-KUMAR ET AL.
sciences, geoscience research can act as a link between progress achieved in basic science and technology that can effectively transfer information for society. Henceforth, the keyword for the future is ‘Earth system science’, where natural science can sustain life and vice versa. Antarctica is an ideal place for resolving the complex problems of geoscience studies. MS-K acknowledges grant (No. 18740319) from the Ministry of Education, Culture, Sports, Science and Technology, Japan. Many of the ideas presented in this paper have evolved through discussions in symposia and informal meetings. The authors, therefore, cannot single out any individuals who might have influenced the contents. We thank the geoscientific community as a whole. The motive behind this paper was to explore new avenues for future geosciences research in the Antarctic continent; we suspect that we have hardly skimmed the topic. We owe gratitude to the numerous scientists who have taken part in JARE expeditions over the past 50 years. Readers who are interested in obtaining more information on JARE activities may visit the website http://ci.nii.ac.jp/organ/journal/ INT1000001377_en.html, where publications of the National Institute of Polar Research, Tokyo, are available for open access. We thank B. Pankhurst and B. Thomas for their constructive comments which helped improve the style and content of this contribution. Finally, the present paper would not have seen light without the constant support and encouragement to the senior author from the co-editors of the volume, K. Shiraishi, Y. Motoyoshi, Y. Hiroi and Y. Osanai, and in particular P. Leat, the Geological Society of London editor-in-charge of the volume. We express our sincere gratitude to all of them.
References A JISHI , H., K AWANO , Y., K AWAKAMI , T. & I KEDA , T. 2004. Geochronological study of post metamorphic granite from Kasumi Rock, Lu¨tzow-Holm Complex, East Antarctica. Polar Geoscience, 17, 35–44. A SAMI , M., S UZUKI , K. & A DACHI , M. 1997. Th, U and Pb data and CHIME dating of monazites from metamorphic rocks of the Rayner, Lu¨tzow-Holm, Yamato– Belgica and Sør Rondane Complexes, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 10, 130– 152. A SAMI , M., S UZUKI , K. & G REW , E. S. 2005. Monazite and zircon dating by the chemical Th–U– total Pb isochron method (CHIME) from Alashyev Bight to the Sør Rondane Mountains, East Antarctica: a reconnaissance study of the Mozambique suture in eastern Dronning Maud Land. Journal of Geology, 113, 59–82. A SAMI , M., G REW , E. S. & M AKIMOTO , H. 2007. Relict sapphirine þ kyanite and spinel þ kyanite associations in pyrope garnet from the eastern Sør Rondane Mountains, East Antarctica. Lithos, 93, 107– 125. B ISNATH , A., F RIMMEL , H. E., A RMSTRONG , R. A. & B OARD , W. S. 2006. Tectonothermal evolution of the Maud Belt: New SHRIMP U– Pb zircon data
from Gjelsvikfjella, Dronning Maud Land, East Antarctica. Precambrian Research, 150, 95–121. B LACK , L. P., W ILLIAMS , I. S. & C OMPSTON , W. 1986. Four zircon ages from one rock: the history of a 3,930 Ma-old granulite from Mount Sones, Enderby Land, Antarctica. Contributions to Mineralogy and Petrology, 94, 427– 437. B OGER , S. D. & W ILSON , C. J. L. 2005. Early Cambrian crustal shortening and a clockwise P –T –t path from the southern Prince Charles Mountains, East Antarctica: implications for the formation of Gondwana. Journal of Metamorphic Geology, 23, 603–623. B OGER , S. D., W ILSON , C. J. L. & F ANNING , C. M. 2001. Early Palaeozoic tectonism within the East Antarctic craton: the final suture between east and west Gondwana? Geology, 29, 463– 466. B OWRING , S. A., W ILLIAMS , I. S. & C OMPSTON , W. 1989. 3.96 Ga gneisses from the Slave province, Northwest Territories, Canada. Geology, 17, 971– 975. B ROWN , M. 2007. Metamorphic conditions in orogenic belts: a record of secular change. International Geological Reviews, 49, 193–234. C ARSON , C. J., A GUE , J. J. & C OATH , C. D. 2002. U –Pb geochronology from Tonagh Island, East Antarctica: implications for the timing of ultra-high temperature metamorphism of the Napier Complex. Precambrian Research, 116, 237– 263. C AVOSIE , A. J., V ALLEY , J. W., W ILDE , S. A. & E DINBURGH ION MICROPROBE FACILITY (EIMF) 2005. Magmatic d18O in 4400– 3900 Ma detrital zircons: A record of the alteration and recycling of crust in the Early Archaean. Earth and Planetary Science Letters, 235, 663–684. D ALLMANN , W. K., A USTRHEIM , H., B UCHER N URMINEN , K. & O HTA , Y. 1990. Geology around the Norwegian Antarctic station ‘Troll’, Jutulsessen, Dronning Maud Land. Norsk Polarinstitutt Meddelelser, 111. D ALZIEL , I. 1991. Pacific margins of Laurentia and East Antarctica–Australia as a conjugate rift pair: evidence and implications for an Eocambrian supercontinent. Geology, 19, 598– 601. D OBMEIER , C. J. & R AITH , M. M. 2003. Crustal architecture and evolution of the Eastern Ghats Belt and adjacent regions of India. In: Y OSHIDA , M., W INDLEY , B. F. & D ASGUPTA , S. (eds) Proterozoic East Gondwana: Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 145–168. D UNKLEY , D. J. 2007. Isotopic zonation in zircon as a recorder of progressive metamorphism. Geochimica et Cosmochimica Acta, 71, A244. E VANS , D. A. D. 2000. Stratigraphic, geochronological and paleomagnetic constraints upon the Neoproterozoic glacial paradox. American Journal of Science, 300, 347– 433. F ANNING , C. M., M OORE , D. H., B ENNETT , V. C. & D ALY , S. J. 1996. The ‘Mawson Continent’: Archaean to Proterozoic crust in the East Antarctic Shield and Gawler Craton, Australia. A cornerstone in Rodinia and Gondwanaland. In: 14th Australian Geological Convention, Geological Society of Australia, Abstracts Volume No. 41, 135.
GEOSCIENCES RESEARCH IN EAST ANTARCTICA F ITZSIMONS , I. C. W. 2000. Grenville-age basement provinces in East Antarctica; evidence for three separate collisional orogens. Geology, 28, 879– 882. F RASER , G. 1997. Geochronological constraints on the metamorphic evolution and exhumation of the Lu¨tzow Holm Complex, East Antarctica. PhD, thesis, The Australian National University, Canberra, A.C.T. F RASER , G., M C D OUGALL , I., E LLIS , D. J. & W ILLIAMS , I. S. 2000. Timing and rate of isothermal decompression in Pan-African granulites from Rundva˚gshetta, East Antarctica. Journal of Metamorphic Geology, 18, 441– 454. G OLYNSKY , A. V. 2007. Magnetic anomalies in East Antarctica and surrounding regions: a window on major tectonic provinces and their boundaries. In: C OOPER , A. K., R AYMOND , C. R. & ISAES Editorial Team (eds) Proceedings of the 10th ISAES. US Geological Survey Open-File Report, 2007-1047, Short Research Paper 006, doi:10.3133/of20071047.srp006. G OLYNSKY , A. & J ACOBS , J. 2001. Grenville-Age versus Pan-African Magnetic Anomaly imprints in Western Dronning Maud Land, East Antarctica. Journal of Geology, 109, 136– 142. G OLYNSKY , A. V., A LYAVDIN , S. V., M ASOLOV , V. N., T SCHERINOV , A. S. & V OLNUKHIN , V. S. 2002. The composite magnetic anomaly map of the East Antarctica. Tectonophysics, 347, 109– 120. G RANTHAM , G. H., M ABOKO , M. & E GLINGTON , B. M. 2003. A review of the evolution of the Mozambique Belt and implications for the amalgamation of Rodinia and Gondwana. In: Y OSHIDA , M., W INDLEY , B. F. & D ASGUPTA , S. (eds) Proterozoic East Gondwana, Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 401–426. G RAY , D. R., F OSTER , D. A., M EERT , J. G., G OSCOMBE , B. D., A RMSTRONG , R., T ROUW , R. A. J. & P ASSCHIER , C. W. 2008. A Damaran perspective on the assembly of southwestern Gondwana. In: P ANKHURST , R. J., T ROUW , R. A. J., D E B RITO N EVES , B. B. & D E W IT , M. J. (eds) West Gondwana: Pre-Cenozoic Correlations across the South Atlantic Region. Geological Society, London, Special Publications, 294, 257–278. G REW , E. S., M ANTON , W. I., A SAMI , A. & M AKIMOTO , H. 1992. Reconnaissance geochronologic data on Proterozoic polymetamorphic rocks of the eastern Sør Rondane Mountains, East Antarctica. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctica Earth Science. Terra, Tokyo, 37–44. H ALPIN , J. A., G ERAKITEYS , C. L., C LARKE , G. L., B ELOUSOVA , E. A. & G RIFFIN , W. L. 2005. In-situ U–Pb geochronology and Hf isotope analyses of the Rayner Complex East Antarctica. Contributions to Mineralogy and Petrology, 148, 689–706. H ALVERSON , G. P., H OFFMAN , P. F., S CHRAG , D. P., M ALOOF , A. C. & R ICE , A. H. N. 2005. Towards a Neoproterozoic composite carbon isotope record. Geological Society of America Bulletin, 117, 1181–1207. H ARLEY , S. L. 1987. A pyroxene-bearing metaironstone and other pyroxene-granulites from Enderby Land, Antarctica: Further evidence for very high temperature
17
(.980 8C) Archaean regional metamorphism in the Napier Complex. Journal of Metamorphic Geology, 5, 341–356. H ARLEY , S. L. 2004. Extending our understanding of ultrahigh temperature crustal metamorphism. Journal of Mineralogical and Petrological Sciences, 99, 140– 158. H ARLEY , S. L. & B LACK , L. P. 1997. A revised Archaean chronology for the Napier Complex, Enderby Land, from SHRIMP ion-microprobe studies. Antarctic Science, 9, 74– 91. H ARLEY , S. L. & H ENSEN , B. J. 1990. Archaean and Proterozoic high-grade terranes of East Antarctica (40–808E): A case study of diversity in granulite facies metamorphism. In: A SHWORTH , J. R. & B ROWN , M. (eds) High Temperature Metamorphism and Crustal Anatexis. Unwin Hyman, London, 320– 370. H ARLEY , S. L. & M OTOYOSHI , Y. 2000. Al zoning in orthopyroxene in a sapphirine quartzite: evidence for .1120 8C UHT metamorphism in the Napier Complex, Antarctica, and implications for the entropy of sapphirine. Contributions to Mineralogy and Petrology, 138, 293– 307. H ARRISON , T. M. & W ATSON , E. B. 1983. Kinetics of zircon dissolution and zirconium diffusion in granitic melts with variable water content. Contributions to Mineralogy and Petrology, 84, 66–72. H IROI , Y., S HIRAISHI , K., Y ANAI , K. & K IZAKI , K. 1983. Aluminum silicates in the Prince Olav and Soya Coasts, East Antarctica. Memoirs of National Institute of Polar Research Special Issue, 28, 115– 131. H IROI , Y., S HIRAISHI , K., M OTOYOSHI , Y., K ANISAWA , S., Y ANAI , K. & K IZAKI , K. 1986. Mode of occurrence, bulk chemical compositions, and mineral textures of ultramafic rocks in the Lu¨tzow-Holm Complex, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 43, 62–84. H IROI , Y., S HIRAISHI , K., M OTOYOSHI , Y. & K ATSUSHIMA , T. 1987. Progressive metamorphism of calc-silicate rocks from the Prince Olav and Soya Coasts, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 1, 73–97. H IROI , Y., S HIRAISHI , K. & M OTOYOSHI , Y. 1991. Late Proterozoic paired metamorphic complexes in East Antarctica, with special reference to the tectonic significance of ultramafic rocks. In: T HOMSON , M. R. A., C RAME , J. A. & T HOMSON , J. W. (eds) Geological Evolution of Antarctica. Cambridge University Press, Cambridge, 83– 87. H IROI , Y., M OTOYOSHI , Y., S ATISH -K UMAR , M., K AGASHIMA , S., S UDA , Y. & I SHIKAWA , N. 2006. Granulites from Cape Hinode in the amphibolitefacies eastern part of Prince Olav Coast, East Antarctica: New evidence for allochthonous block in the Lu¨tzow-Holm Complex. Polar Geoscience, 19, 89–108. H OFFMAN , P. F. 1991. Did the breakout of Laurentia turn Gondwanaland inside out? Science, 252, 1409–1412. H OFFMAN , P. F., K AUFMAN , A. J., H ALVERSON , G. P. & S CHRAG , D. P. 1998. A Neoproterozoic snowball Earth. Science, 281, 1342– 1346. H OKADA , T. 2001. Feldspar thermometry in ultrahightemperature metamorphic rocks: evidence of crustal
18
M. SATISH-KUMAR ET AL.
metamorphism attaining 1100 8C in the Archaean Napier Complex, East Antarctica. American Mineralogist, 86, 932–938. H OKADA , T. & M OTOYOSHI , Y. 2006. Electron microprobe technique for U– Th–Pb and REE chemistry of monazite, and its implications for pre-, peak- and post-metamorphic events of the Lu¨tzow-Holm Complex and the Napier Complex, East Antarctica. Polar Geoscience, 19, 118–151. H OKADA , T. & S UZUKI , S. 2006. Feldspar in felsic orthogneiss as indicator for UHT crustal processes. Journal of Mineralogical and Petrological Sciences, 101, 260–264. H OKADA , T., M ISAWA , K., S HIRAISHI , K. & S UZUKI , S. 2003. Mid to late Archaean (3.3– 2.5 Ga) tonalitic crustal formation and high-grade metamorphism at Mt. Riiser-Larsen, Napier Complex, East Antarctica. Precambrian Research, 127, 215– 228. I KEDA , T. & K AWAKAMI , T. 2004. Structural analysis of the Lu¨tzow-Holm Complex in Akarui Point, East Antarctica, and overview of the complex. Polar Geoscience, 17, 22– 34. I SHIKAWA , M., M OTOYOSHI , Y., F RASER , G. L. & K AWASAKI , T. 1994. Structural evolution of Rundva˚gshetta region, Lu¨tzow-Holm Bay, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 7, 69– 89. I SHIKAWA , M., H OKADA , T., I SHIZUKA , H., M IURA , H., S UZUKI , S. & T AKADA , M. 2000. Geological map of Mt. Riiser-Larsen, Enderby Land, Antarctica. Antarctic Geological Map Series Sheet 37. National Institute of Polar Research, Tokyo. I SHIKAWA , T. 1976. Superimposed folding of the Precambrian metamorphic rocks of the Lu¨tzow-Holm Bay region, East Antarctica. Memoirs of National Institute of Polar Research, Series C, 9, 1–41. I SHIZUKA , H., S UZUKI , S. & N AKAMURA , A. 2002. Peak temperatures of ultra-high temperature metamorphism of the Napier Complex, Enderby Land, East Antarctica, as deduced from porphyroclastic pyroxenes of meta-ultramafic rocks. Polar Geoscience, 15, 1 –16. J ACOBS , J. & T HOMAS , R. J. 2004. A Himalayan-type indenter-escape tectonic model for the southern part of the Late Neoproterozoic/Early Palaeozoic East African–Antarctic Orogen. Geology, 32, 721– 724. J ACOBS , J., F ANNING , C. M., H ENJES -K UNST , F., O LESH , M. & P AECH , H.-J. 1998. Continuation of the Mozambique Belt into East Antarctica: Grenville age metamorphism and polyphase Pan-African high grade events in Central Dronning Maud Land. Journal of Geology, 106, 385–406. J ACOBS , J., F ANNING , C. M. & B AUER , W. 2003a. Timing of Grenville-age vs. Pan-African medium- to high grade metamorphism in western Dronning Maud Land (East Antarctica) and significance for correlations in Rodinia and Gondwana. Precambrian Research, 125, 1– 20. J ACOBS , J., K LEMD , J. R., F ANNING , C. M., B AUER , W. & C OLOMBO , F. 2003b. Extensional collapse of the Late Neoproterozic –Early Palaeozoic East African– Antarctic Orogen in central Dronning Maud Land, East Antarctica. In: Y OSHIDA , M.,
W INDLEY , B. F. & D ASGUPTA , S. (eds) Proterozoic East Gondwana: Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 271– 287. J OHANNES , W. & H OLTZ , F. 1996. Petrogenesis and Experimental Petrology of Granitic Rocks. Springer, Berlin. K AMENEV , E. N. 1972. Geological structure of Enderby Land. In: A DIE , R. J. (ed.) Antarctic Geology and Geophysics. Universitetsforlaget, Oslo, 579–583. K AMENEV , E. N. 1975. The geology of Enderby Land. Academy of Science of the USSR Committee on Antarctic Research Report, 14. K AMENEV , E. N. 1991. Structure and evolution of the Antarctic shield in Precambrian. In: F INDLAY , R. H., U NRUG , R., B ANKS , M. R. & V EEVERS , J. J. (eds) Gondwana Eight. Balkema, Rotterdam, 141–151. K AWAKAMI , T. & I KEDA , T. 2004a. Structural evolution of the Ongul Islands, Lu¨tzow-Holm Complex, East Antarctica. Polar Geoscience, 17, 12–21. K AWAKAMI , T. & I KEDA , T. 2004b. Timing of ductile deformation and peak metamorphism in Skallevikshalsen, Lu¨tzow-Holm Complex, East Antarctica. Polar Geoscience, 17, 1– 11. K AWAKAMI , T. & M OTOYOSHI , Y. 2004. Timing of attainment of spinel þ quartz coexistence in garnet– sillimanite leucogneiss from Skallevikshalsen, Lu¨tzowHolm Complex, East Antarctica. Journal of Mineralogical and Petrological Sciences, 99, 311–319. K AWAKAMI , T., E LLIS , D. J. & C HRISTY , A. G. 2006. Sulfide evolution in high-temperature to ultrahightemperature metamorphic rocks from Lu¨tzow-Holm Complex, East Antarctica. Lithos, 92, 431–446. K AWASAKI , T., I SHIKAWA , M. & M OTOYOSHI , Y. 1993. A preliminary report on cordierite-bearing assemblages from Rundva˚gshetta region, Lu¨tzow-Holm Bay, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 6, 47–56. K ELLY , N. M. & H ARLEY , S. L. 2005. An integrated microtextural and chemical approach to zircon geochronology: refining the Archaean history of the Napier Complex, East Antarctica. Contributions to Mineralogy and Petrology, 149, 57–84. K ELLY , N. M., C LARKE , G. L. & F ANNING , C. M. 2002. A two-stage evolution of the Neoproterozoic Rayner Structural Episode: new U –Pb sensitive high resolution ion microprobe constraints from the Oygarden Group, Kemp Land, East Antarctica. Precambrian Research, 116, 307– 330. K ELSEY , D. E., W ADE , B. P., C OLLINS , A. S., H AND , M., S EALING , C. R. & N ETTING , A. 2008. Discovery of a Neoproterozoic basin in the Prydz belt in East Antarctica and its implications for Gondwana assembly and ultrahigh temperature metamorphism. Precambrian Research, 161, 355– 388. K EMP , A. I. S., H AWKESWORTH , C. J., P ATERSON , B. A. & K INNY , P. D. 2006. Episodic growth of the Gondwana supercontinent from hafnium and oxygen isotopes in zircon. Nature, 439, 580–583. K IZAKI , K. 1962. Structural geology and petrology of East Ongul Island, East Antarctica. Part I. Structural geology. Nankyoku Shiryoˆ (Antarctic Records), 14, 27–35. K IZAKI , K. 1964. Tectonics and petrography of the East Ongul Island, Lu¨tzow Holm Bukt, East
GEOSCIENCES RESEARCH IN EAST ANTARCTICA Antarctica. JARE Science Reports, Series C (Geology), 2, 1 –24. L AWVER , L. A., G AHAGAN , L. M. & D ALZIEL , I. W. D. 1998. A tight-fit Early Mesozoic Gondwana, a plate reconstructive perspective. Memoirs of National Institute of Polar Research, Special Issue, 53, 214–229. M ARUYAMA , S. & S ANTOSH , M. 2008. Models on Snowball Earth and Cambrian Explosion: A synopsis. Gondwana Research, 14, 22–32. M ATSUMOTO , Y., Y OSHIDA , M. & Y ANAI , K. 1979. Geology and geologic structure of the Langhovde and Skarvsnes regions, East Antarctica. Memoirs of National Institute for Polar Research, Special Issue, 14, 106–120. M ATSUMOTO , Y., N ISHIDA , T., Y ANAI , K. & K OJIMA , H. 1982. Geology and geologic structure of the northern Ongul Islands and surroundings, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 21, 47– 70. M EERT , J. G. 2003. A synopsis of events related to the assembly of eastern Gondwana. Tectonophysics, 362, 1–40. M EERT , J. G. 2007. Testing the Neoproterozoic glacial models. Gondwana Research, 11, 573–574. M EERT , J. G. & L IEBERMAN , B. S. 2008. The Neoproterozoic assembly of Gondwana and its relationship to the Ediacaran– Cambrian radiation. Gondwana Research, 14, 5– 21. M EZGER , K. & C OSCA , M. A. 1999. The thermal history of the Eastern Ghats Belt (India) as revealed by U– Pb and 40Ar/39Ar dating of metamorphic and magmatic minerals: implications for the SWEAT correlation. Precambrian Research, 94, 251– 271. M IKHALSKY , E. V., S HERATON , J. W., L AIBA , A. A., T INGEY , R. J., T HOST , D. E., K AMENEV , E. N. & F EDOROV , L. V. 2001. Geology of the Prince Charles Mountains, Antarctica. AGSO Bulletin, Canberra, 247. M IKHALSKY , E. V., B ELIATSKY , B. V., S HERATON , J. W. & R OLAND , N. W. 2006. Two distinct Precambrian terranes in the Southern Prince Charles Mountains, East Antarctica: SHRIMP dating and geochemical constraints. Gondwana Research, 9, 291– 309. M ICHIBAYASHI , K., O KAMOTO , A., M ASUZAWA , T., K AWAKAMI , T., I KEDA , T. & Y ASUDA , H. 2004. Orientation contrast images of garnet in granulitefacies quartzite, Lu¨tzow-Holm Complex, East Antarctica. Journal of Geological Society of Japan, 110, V– VI. M OTOYOSHI , Y. 1986. Prograde and progressive metamorphism of the granulite facies Lu¨tzow-Holm Bay region, East Antarctica. DSc thesis, Hokkaido University, Sapporo. M OTOYOSHI , Y. & I SHIKAWA , M. 1997. Metamorphic and structural evolution of granulites from Rundva˚gshetta, Lu¨tzow-Holm Bay, East Antarctica. In: R ICCI , C. A. (ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antartica, Sienna, 65– 72. M OTOYOSHI , Y., H OKADA , T., H IROI , Y. & S HIRAISHI , K. 2004. EMP dating on Cape Hinode of the Lu¨tzow-Holm Complex, East Antarctica. Abstracts of 24th Symposium on Antarctic Geoscience, 23.
19
M OTOYOSHI , Y., H OKADA , T. & S HIRAISHI , K. 2006. Electron microprobe (EMP) dating on monazite from Forefinger Point granulites, East Antarctica. In: F U¨ TTERER , D. K., D AMASKE , D., K LEINSCHMIDT , G., M ILLER , H. & T ESSENSOHN , F. (eds) Antarctica, Contributions to Global Earth Sciences. Springer, Berlin, 63– 68. M U¨ LLER , W. 2003. Strengthening the link between geochronology, textures and petrology. Earth and Planetary Science Letters, 206, 237– 251. N ISHI , N., K AWANO , Y. & K AGAMI , H. 2002. Rb– Sr and Sm–Nd isotopic geochronology of the granitoid and hornblende biotite gneiss from Oku-iwa Rock in the Lu¨tzow-Holm Complex, East Antarctica. Polar Geoscience, 15, 46– 65. N OGI , Y., K ITADA , K. & S EAMA , N. 2006. Seafloor geophysical surveys during JARE-47. Abstracts of 26th Symposium on Antarctic Geoscience, 23. O HTA , Y. 1999. Nature environment map, Gjelsvikfjella and Western Mu¨hlig-Hofmannfjella, sheets 1 and 2, Dronning Maud Land. Norsk Polarinstitutt, Temakart 24. O KAMOTO , A. & M ICHIBAYASHI , K. 2005. Progressive shape evolution of a mineral inclusion under differential stress at high temperature: example of garnet inclusions within a granulite-facies quartzite from the Lu¨tzow-Holm Complex, East Antarctica. Journal of Geophysical Research, 110, B11203, doi:10.1029/2004JB003526. O SANAI , Y., S HIRAISHI , K., T AKAHASHI , Y. ET AL . 1992. Geochemical characteristics of the metamorphic rocks from the Central Sør Rondane Mountains, East Antarctica. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Science, Terra, Tokyo, 17–27. O SANAI , Y., T OYOSHIMA , T., O WADA , M. ET AL . 2001. Geological map of Tonagh Island, Enderby Land, Antarctica. Antarctic Geological Map Series Sheet 38. National Institute of Polar Research, Tokyo. O SANAI , Y., T OYOSHIMA , T., O WADA , M. ET AL . 2004. Geological map of Skallen, East Antarctica (Revised version). Antarctic Geological Map Series, Sheet 39. National Institute of Polar Research, Tokyo. P AULSSON , O. & A USTRHEIM , H. 2003. A geochronological and geochemical study of rocks from Gjelsvikfjella, Dronning Maud Land, Antarctica— implications for Mesoproterozoic correlations and assembly of Gondwana. Precambrian Research, 125, 113– 138. P HILLIPS , G., W ILSON , C. J. L., C AMPBELL , I. H. & A LLEN , C. M. 2006. U– Th–Pb detrital zircon geochronology from the southern Prince Charles Mountains East Antarctica—defining the Archaean to Neoproterozoic Ruker Province. Precambrian Research, 148, 292–306. R ODIONOV , N. V., B ELYATSKY , B. V., M ATUKOV , D. I., A NTONOV , A. V., P RESNYAKOV , S. L., L EITCHENKOV , G. L. & S ERGEEV , S. A. 2006. First SHRIMP U– Th–Pb data on detrital zircons and monazite from ice drillcore, Vostok lake, Antarctica. Geochimica et Cosmochimica Acta, 70, A536. R UBATTO , D. 2002. Zircon trace element geochemistry: partitioning with garnet and the link between U –Pb ages and metamorphism. Chemical Geology, 184, 123– 138.
20
M. SATISH-KUMAR ET AL.
S ATISH -K UMAR , M., H ERMANN , J., T SUNOGAE , T. & O SANAI , Y. 2006a. Carbonation of Cl-rich scapolite boudins in Skallen, East Antarctica: evidence for changing fluid condition in the continental crust. Journal of Metamorphic Geology, 24, 241–261. S ATISH -K UMAR , M., K AGASHIMA , S., S UDA , Y. & M OTOYOSHI , Y. 2006b. Geology of Byobu Rock and Gobanme Rock, Prince Olav Coast, East Antarctica. Polar Geoscience, 19, 1 –36. S HERATON , J. W., T INGEY , R. J., B LACK , L. P., O FFE , L. A. & E LLIS , D. J. 1987. Geology of Enderby Land and western Kemp Land, Antarctica. Bulletin, Australian Bureau of Mineral Resources, 223, 1– 51. S HIMURA , T., F RASER , G., T SUCHIYA , N. & K AGAMI , H. 1998. Genesis of migmatites of Bradvagnipa, East Antarctica. Memoirs of the National Institute of Polar Research, Special Issue, 53, 109– 136. S HIRAISHI , K., E LLIS , D. J., F ANNING , C. M., H IROI , Y., K AGAMI , H. & M OTOYOSHI , Y. 1997. Re-examination of the metamorphic and protolith ages of the Rayner Complex, Antarctica: evidence for the Cambrian (Pan-African) regional metamorphic event. In: R ICCI , C. A. (ed.) The Antarctic Region: Geological Evolution and Processes, Terra Antartica, Sienna, 79–88. S HIRAISHI , K., H OKADA , T., F ANNING , C. M., M ISAWA , K. & M OTOYOSHI , Y. 2003. Timing of thermal events in eastern Dronning Maud Land, East Antarctica. Polar Geoscience, 16, 76–99. S HIRAISHI , K., A SAMI , M., I SHIZUKA , H. ET AL . 1991. Geology and metamorphism of the Sør Rondane Mountains, East Antarctica. In: T HOMSON , M. R.A., C RAME , J. A. & T HOMSON , J. W. (eds) Geological Evolution of Antactica. Cambridge University Press, Cambridge, 77–82. S HIRAISHI , K., H IROI , Y., E LLIS , D. J., F ANNING , C. M., M OTOYOSHI , Y. & N AKAI , Y. 1992. The first report of a Cambrian orogenic belt in East Antarctica—an ion microprobe study of the Lu¨tzow-Holm Complex. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Science. Terra, Tokyo, 67– 73. S HIRAISHI , K., E LLIS , D. J., H IROI , Y., F ANNING , C. M., M OTOYOSHI , Y. & N AKAI , Y. 1994. Cambrian orogenic belt in East Antarctica and Sri Lanka: Implications for Gondwana assembly. Journal of Geology, 102, 47–65. S HIRAISHI , K., H OKADA , T., F ANNING , C. M., M ISAWA , K. & M OTOYOSHI , Y. 2003. Timing of
thermal events in the Dronning Maud Land, East Antarctica. Polar Geoscience, 16, 76–99. S TERN , R. J. 2008. Neoproterozoic crustal growth: The solid Earth system during a critical episode of Earth history. Gondwana Research, 14, 33–50. S UDA , Y., K AGASHIMA , S., S ATISH -K UMAR , M., M OTOYOSHI , Y. & H IROI , Y. 2006. Geochemistry of mafic metamorphic rocks in the Lu¨tzow-Holm Complex, East Antarctica: implications for tectonic evolution. Polar Geoscience, 19, 62–88. V ANCE , D., M ULLER , W. & V ILLA , I. M. 2003. Geochronology: linking the isotopic record with petrology and textures—an introduction. In: V ANCE , D., M U¨ LLER , W. & V ILLA , I. (eds) Geochronology: Linking the Isotope Record with Petrology and Textures. Geological Society, London, Special Publications, 220, 1 –24. VON F RESE , R. R. B., G OLYNSKY , A. V., K IM , H. R., ET AL . 2007. The next generation Antarctic Digital Magnetic Anomaly Map (ADMAP). In: C OOPER , A. K., R AYMOND , C. R. ISAES Editorial Team (eds) Proceedings of the 10th ISAES. US Geological Survey Open-File Report, 2007-1047, Short Research Paper 093, doi:10.3133/of2007-1047.srp093. W ATSON , E. B. & H ARRISON , T. M. 1983. Zircon saturation revisited: temperature and composition effects in a variety of crustal magma types. Earth and Planetary Science Letters, 64, 295– 304. Y OSHIDA , M. 1977. Geology of the Skallen region, Lu¨tzow-Holmbutka, East Antarctica. Memoirs of National Institute of Polar Research, Series C (Earth Science), 11, 1–38. Y OSHIDA , M. 1978. Tectonics and petrology of charnockites around Lu¨tzow-Holmbukta, East Antarctica. Journal of Geosciences, Osaka City University, 21, 65–152. Y OSHIDA , M. 1992. Late Proterozoic to early Palaeozoic events in East Gondwanan crustal fragments. In: 29th International Geological Congress, Kyoto, Japan, Abstracts, 265. M. 2007. Geochronological data Y OSHIDA , evaluation: implications for the Proterozoic tectonics of East Gondwana. Gondwana Research, 12, 228–241. Y OUNG , D. N. & B LACK , L. P. 1991. U–Pb zircon dating of Proterozoic igneous charnockites from the Mawson Coast, East Antarctica. Antarctic Science, 3, 205–216.
Geochronological constraints on the Late Proterozoic to Cambrian crustal evolution of eastern Dronning Maud Land, East Antarctica: a synthesis of SHRIMP U – Pb age and Nd model age data KAZUYUKI SHIRAISHI1,2, DANIEL J. DUNKLEY1, TOMOKAZU HOKADA1,2, C. MARK FANNING3, HIROO KAGAMI4 & TAKUJI HAMAMOTO4,5 1
National Institute of Polar Research, 1-9-10, Kaga, Itabashi-ku, Tokyo 173-8515, Japan (e-mail:
[email protected]) 2
Department of Polar Studies, Graduate School of Advanced Studies (SOKENDAI), Tokyo 173-8515, Japan
3
Research School of Earth Sciences, The Australian National University, A.C.T 0200, Australia 4
Graduate School of Science and Technology, Niigata University, Niigata 950-2181, Japan
5
Present address: Dia Consultants Co. Ltd, Kanayama-cho 1-6-12, Nagoya 456-0002, Japan Abstract: In eastern Dronning Maud Land (DML), East Antarctica, there are several discrete, isolated magmatic and high-grade metamorphic regions. These are, from west (c. 208E) to east (c. 508E), the Sør Rondane Mountains (SRM), Yamato– Belgica Complex (YBC), Lu¨tzowHolm Complex (LHC), Rayner Complex (RC) and Napier Complex (NC). To understand this region in a Gondwanan context, one must distinguish between Pan-African and Grenvillian aged magmatic and metamorphic events. Sensitive high-resolution ion microprobe U –Pb zircon ages and Nd model ages for metamorphic and plutonic rocks are examined in conjunction with published geological and petrological studies of the various terranes. In particular, the evolution of the SRM is examined in detail. Compilation of Nd model ages for new and published data suggests that the main part of eastern Dronning Maud Land, including the SRM, represents juvenile late Mesoproterozoic (c. 1000–1200 Ma) crust associated with minor fragments of an older continental component. Evidence for an Archaean component in the basement of the SRM is lacking. As for central DML, 1100– 1200 Ma extensive felsic magmatism is recognized in the SRM. Deposition of sediments during or after magmatism and possible metamorphism at 800–700 Ma is recognized from populations of detrital zircon in metasedimentary rocks. The NE Terrane of the SRM, along with the YBC, was metamorphosed under granulite-facies conditions at c. 600–650 Ma. The SW and NE Terranes of the SRM were brought together during amphibolite-facies metamorphism at c. 570 Ma, and share a common metamorphic and magmatic history from that time. High-grade metamorphism was followed by extensive A-type granitoid activity and contact metamorphism between 560 and 500 Ma. In contrast, TDM and inherited zircon core ages suggest that the LHC is a collage of protoliths with a variety of Proterozoic and Archaean sources. Later peak metamorphism of the LHC at 520–550 Ma thus represents the final stage of Gondwanan amalgamation in this section of East Antarctica.
The eastern part of Dronning Maud Land (DML) between 378E and 508E, which represents the original definition of Dronning Maud Land as a whole, was discovered by the Norwegian aviator Riiser-Larsen in 1930. Most of the coastal and inland outcrops were air-photographed by the Lars Christensen Expedition in 1937 and by Operation Highjump in 1946– 1947. For the International Geophysical Year (IGY) of 1957, the Japanese Antarctic Research Expedition (JARE) established the first wintering station in East Ongul Island, located in the mouth of Lu¨tzow-Holm Bay. Since then, the footprints of geologists have covered eastern DML from 208E to 508E.
Prior to JARE, the Soviet Antarctic Expedition started to investigate an extensive area of the Indian Sector of the East Antarctica and the Australian Antarctic Research Expedition (ANARE) worked eastwards from 458E. In particular, ANARE performed a large-scale geological investigation of Enderby Land between 1974 and 1980. Belgium started wintering during IGY on Princess Ragnhild Coast and conducted pioneer investigations in the Sør Rondane Mountains until 1961. Amongst the various national expeditions, geologists from many countries have participated in field programmes in eastern DML and Enderby Land.
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 21– 67. DOI: 10.1144/SP308.2 0305-8719/08/$15.00 # The Geological Society of London 2008.
22
K. SHIRAISHI ET AL.
The results of a decade of geological surveys following IGY were compiled in the Antarctic Map Folio Series (Bushnell & Craddock 1969, 1970). The early stages of field investigations by JARE were devoted to geological mapping, resulting in 36 sheets of geological maps on the scales of 1:5000 to 1:250 000, published as the Antarctic Geological Map Series by the National Institute of Polar Research, Tokyo. Reviews of the basement geology of East Antarctica based on the studies before the 1980s were published by Ravich & Kamenev (1975), Grew (1982) and Tingey (1991). Since then, the accumulation of precise age data by thermal ionization mass spectrometry (TIMS) and sensitive high-resolution ion microprobe (SHRIMP), detailed petrological studies by electron probe micro analysis (EPMA) and intense field investigations have made it possible to describe the crustal evolution of East Antarctica in relation to the formation of Gondwana (e.g. Fitzsimons 2000; Yoshida et al. 2003; Frimmel 2004; Jacobs & Thomas 2004). In particular, studies on Dronning Maud Land have focused on continental plate reconstructions, and this region is now widely regarded to be the southern continuation of the late Neoproterozoic to early Palaeozoic East African Orogen (e.g. Jacobs et al. 1998, 2003a, b; Jacobs & Thomas 2002; Grantham et al. 2003; Paulsson & Austrheim 2003). In eastern Dronning Maud Land and adjacent areas, there are several discrete, isolated exposures
of magmatic and high-grade metamorphic terranes. These are, from west (c. 208E) to east (c. 508E), the Sør Rondane Mountains (SRM), the Yamato – Belgica Complex (YBC), the Lu¨tzow-Holm Complex (LHC), the Rayner Complex (RC) and the Napier Complex (NC) (Fig. 1). Hiroi et al. (1991) compiled the geology and petrology of the Lu¨tzow-Holm Complex, which extends along the eastern coastline of Lu¨tzow-Holm Bay and the Prince Olav Coast, and is characterized by a continuous increase in metamorphic grade and a clockwise prograde P – T path. The Yamato – Belgica Complex is characterized by widespread igneous activity and low-P/high-T type metamorphism, for which Hiroi et al. (1991) proposed a tectonic scenario involving continent – continent collision. Petrological studies on the LHC revealed ultrahigh-temperature metamorphism in the southern part of the complex (Motoyoshi & Ishikawa 1997). Shiraishi et al. (1992, 1994, 2003) reported c. 550–530 Ma SHRIMP zircon U –Pb ages for peak metamorphism in the LHC and suggested that the collision took place in the last stage of Gondwana construction during the Pan-African Orogeny (Fig. 2). Since then, Pan-African events (c. 600–500 Ma) involving extensive plutonic activity and high-grade metamorphism have been recognized in many parts of East Antarctica. Whether there was a single Grenville-aged Circum-East Antarctic mobile belt that was tectonically reactivated in Pan-African times (Yoshida et al. 2003) or a Pan-African
Fig. 1. Map of geological terranes in central–eastern Dronning Maud Land to Enderby Land, East Antarctica. The areas of Figures 2 and 3 are marked. NE, NE Terrane; SW, SW Terrane; wRC, Western Rayner Complex.
AGE CONSTRAINTS IN EAST ANTARCTICA
23
Fig. 2. Map of SHRIMP zircon U– Pb ages for rocks from the Lu¨tzow-Holm Complex. i, igneous age; m, metamorphic age (ages in million years). *Ongul Is. includes East Ongul Island, Nesøya, Fleynøya and Utholmen. Modified from Shiraishi et al. (2003).
juxtaposition and assembly of three unrelated Grenville-aged terranes into East Gondwana (Fitzsimons 2000), is a major point of contention. In this context, the correlation of the Pan-African Mozambique Suture with geological features in the SRM to the west of the LHC is a major focus of current research. Recent comprehensive compilations of geological history of East Antarctica reveal the essential involvement of East Antarctica in the tectonic development of Gondwana (e.g. Fitzsimons 2000; Jacobs & Thomas 2004; Meert & Lieberman 2008). In particular, studies indicate that eastern DML is at a critical location in terms of the intersection of the Mozambique belt and various Pan-African terranes from the Prince Charles Mountains in Mac Robertson Land to the Pinjara Orogen in Western Australia (e.g. Jacobs et al. 1998; Fitzsimons 2000; Grantham 2003; Jacobs & Thomas 2004; Bisnath et al. 2006; Mikhalsky et al. 2006; Meert & Lieberman 2007). Boger et al. (2001) proposed an early Palaeozoic orogenic belt extending from Western Australia to Mac
Robertson Land and suggested possible links with the Lu¨tzow-Holm Complex. A similar but more extensive orogenic belt, penetrating through the LHC and across central Africa, was proposed by Meert (2001, 2003), who combined geochronological and palaeomagnetic data to develop a polyphase model for eastern Gondwana assembly. In recent years, two major models have emerged for Neoproterozoic –early Palaeozoic tectonism involved in the formation of East Gondwana: (1) Pan-African structures are coeval over long distances, forming a broad linear orogen from East Antarctica to southern East Africa (East Africa– Antarctic Orogen (EAAO); e.g. Jacobs & Thomas 2002, 2004); (2) two overlapping orogens were involved in the amalgamation of East Gondwana: an older East Africa Orogeny (Stern 1994) and a younger Kuunga Orogeny (Meert 2003). Both models identify two stages of Pan-African orogenesis, at 750–620 Ma and 570–530 Ma (Jacobs et al. 2003b; Meert 2003; Grantham et al. 2008). Although comprehensive geochronological studies have been published on eastern DML, the
24
K. SHIRAISHI ET AL.
timing of key magmatic, deformational and metamorphic events has not been sufficiently resolved to place these events in a tectonic framework related to the formation of Gondwana. Pan-African v. Grenvillian orogenesis is uncertain. In this paper we present zircon and titanite SHRIMP U –Pb ages for 11 metamorphic and magmatic rocks from the Sør Rondane Mountains, the largest mountain range in eastern DML. In addition, new Nd model ages, mainly from the LHC, are compared with published ages from central–eastern Dronning Maud Land to Enderby Land. Geochronological data are examined in conjunction with geological and petrological studies of the respective terranes. The results provide constraints on the Neoproterozoic to Cambrian crustal history of eastern Dronning Maud Land, and present insights into a region that played a critical role in the formation of the Gondwanan supercontinent.
Geological outline of the Sør Rondane Mountains The Sør Rondane Mountains (SRM), located between 228 and 288E and 71.58 and 72.58S, form one of the largest mountain chains in East Antarctica. The SRM are underlain by medium- to highgrade metamorphic rocks with various intrusions of plutonic rocks and minor mafic dykes (e.g. Shiraishi et al. 1991, 1997a). The mountains are divided into the NE and SW Terranes by an inferred tectonic line, the Sør Rondane Suture (SRS: Osanai et al. 1992). The NW Terrane is mainly composed of granulite-facies metamorphic rocks of pelitic, psammitic and intermediate compositions, whereas the SW Terrane is composed of amphibolite-facies and lower-grade metamorphic rocks of mainly intermediate to basic composition, including a large volume of meta-tonalite (Fig. 3). Plutonic rocks and dykes of various sizes intrude metamorphic lithologies in both terranes, and consist of syn- to post-orogenic granite, syenite, diorite and alkaline mafic dykes. Metamorphosed basic and intermediate igneous rocks in the central part of the mountains indicate that magmatic protoliths have geochemical affinities with oceanic, island arc, accretionary complex and continental margin arc settings in modern plate-tectonic systems (Osanai et al. 1992). Ishizuka et al. (1996) reported that ultramafic gneiss in the NE Terrane evolved from a midocean ridge basalt (MORB)-like magma source. Meta-tonalite from the SW Terrane is characterized by relatively high Na2O/K2O, K/Rb, Sr/Y and (La/Yb)N, low CaO/Na2O and low initial Sr isotopic ratios, which have been attributed to magma
genesis from a hot subducting plate (Ikeda & Shiraishi 1998). Metamorphic P–T conditions of gneisses from the NE Terrane have been estimated by several workers (Asami & Shiraishi 1987; Shiraishi & Kojima 1987; Grew et al. 1989; Asami et al. 1990, 1992, 2007; Ishizuka et al. 1995). The NE Terrane was metamorphosed under granulite-facies conditions (800 8C and 7–8 kbar), with subsequent amphibolite-facies retrogression (530 –580 8C and 5.5 kbar) within the kyanite stability field. Subsequently, the rocks were recrystallized at lower pressures, as indicated by the presence of andalusite in certain lithologies (Asami & Shiraishi 1987; Asami et al. 1992, 1993). Recently, Asami et al. (2007) reported relict sapphirine þ kyanite and spinel þ kyanite associations in garnet from the eastern SRM; such assemblages have also been found in granulite-facies rocks from the LHC and Sri Lanka, which are located close to the SRM in the reconstruction of Gondwana (Asami et al. 2007). From these assemblages, the peak conditions of granulite-facies metamorphism was estimated to be 860–895 8C and 12 kbar based on the sapphirine–spinel thermometer and stability of kyanite, although the stability range of the kyanite þ spinel assemblage is not well defined (Asami et al. 2007). In contrast, such high-T metamorphism has not been recognized in the SW Terrane, where gneisses at Vengen ridge and northern Walnumfjelle experienced metamorphism estimated at upper amphibolite-facies conditions, followed by pervasive retrograde metamorphism in association with mylonitization (Shiraishi & Kojima 1987). Since pioneer studies in the 1960s, numerous Rb –Sr and K –Ar geochronological studies in both terranes of the SRM have yielded ages from c. 500 to 420 Ma, attributed to an intense thermal event associated with plutonic activity in the early Palaeozoic (Picciotto et al. 1964; Takahashi et al. 1990; Grew et al. 1992; Shiraishi & Kagami 1992; Shiraishi et al. 1997a). Older mineral and whole-rock isochron ages have also been reported. Shiraishi & Kagami (1992) considered the age of granulite-facies metamorphism in the NE Terrane to be c. 1000 Ma, on the basis of Sm– Nd and Rb –Sr whole-rock isochron ages from orthogneisses. In contrast, internal mineral isochrons from orthogneisses and paragneisses yielded 556 Ma and 624 Ma ages for Rb–Sr and Sm–Nd systems, respectively. These younger ages were attributed to a series of thermal events associated with granitic intrusions. On the basis of Nd model ages, it was also observed that the timing for Grenvillian (c. 1000 Ma) granulite-facies metamorphism requires a short time interval between crustal formation and orogenesis (Grew et al. 1992; Shiraishi & Kagami 1992).
AGE CONSTRAINTS IN EAST ANTARCTICA
Fig. 3. Simplified geological map and sample localities of the Sør Rondane Mountains. SRS, Sør Rondane Suture (Osanai et al. 1992). 25
26
K. SHIRAISHI ET AL.
A radically different interpretation was provided by Asami et al. (1996, 1997, 2005), on the basis of electron microprobe dating of monazite and zircon in granulites from the SRM as well as the LHC, RC and NC. Monazite belonging to granulite-grade metamorphic assemblages, including grains enclosed in garnet porphyroblasts, yielded chemical Th – U – total Pb isochron method (CHIME) ages that mostly range from 550 to 510 Ma. Asami et al. concluded that granulite-facies metamorphism and high-strain deformation took place at 540 – 530 Ma, in a Cambrian mono-metamorphic belt extending from 258E to 458E. It was also inferred that sediments that formed the protoliths of paragneisses were deposited during the Neoproterozoic. In this scenario, c. 1000 Ma ages must date the formation of protoliths, of either igneous or high-grade metamorphic lithologies. Although the monazite dating establishes the widespread significance of high-grade metamorphism during Pan-African orogenesis, it does not clearly address the possibility of multiple metamorphic events, the origins of metamorphic protoliths, or the relationship of deformational structures and magmatic intrusions to the geochronological data. There is also no clear explanation of the differences in metamorphic grade and lithological types between the NE and SW Terranes of the SRM. Sub-grain analysis of zircon by ion microprobe was required to resolve these issues.
SHRIMP U– Pb geochronology of the Sør Rondane Mountains Samples and analytical procedures The present study compiles SHRIMP U – Pb zircon and titanite analyses for 11 samples, obtained during 13 analytical sessions at the SHRIMP II facilities at the Australian National University, Canberra (ANU; sessions A1 – 7) and the National Institute of Polar Research, Tokyo (NIPR; sessions N1 – 6). Data reduction and processing was performed using the Excel add-in program SQUID (v.1.12a; Ludwig 2001) and plots were generated using ISOPLOT (v.3.50; Ludwig 2003). For zircon analysis, abundance of U was calibrated against standard SL13 (238 ppm), and U – Pb measurements were calibrated against 204 Pb-corrected (Pb/U)/(UO/U)2 values for standard AS3 (1099 Ma, Paces & Miller 1993). For each standard dataset, scatter on (Pb/U)/ (UO/U)2 ratios and external spot-to-spot errors are quoted with data from each sample and session in Tables 1 – 14.
All measurements on zircon were corrected for common Pb content using measured 204Pb and a Stacey & Kramers (1975) model for ages approximating those of standard and unknown zircon ages (see Ludwig 2001 for details). The procedure for titanite U – Pb calibration against standard KHAN (700 ppm U and 518 Ma, Kinny et al. 1994) was identical, except that the 207Pb correction for common Pb was used, which assumes concordance between radiogenic 206Pb/238U and 207 Pb/235U ages. Wherever possible, pooled ages were calculated from single analytical sessions using the concordia age function of SQUID, which has the advantage of providing a test of concordance between pooled 206Pb/238U and 207 Pb/206Pb ages. Mean 206Pb/238U ages for pooled data are also provided in Tera – Wasserberg plots. Errors on single spot ratios and ages are quoted at 1s, whereas pooled ages and concordia intercept ages are quoted at 95% confidence levels. Concordia ages are always calculated separately from single sessions with a standard calibration. In some samples, concordia ages for the same generation of zircon growth were obtained in duplicate sessions, to assess the reproducibility of results between the SHRIMP II facilities at ANU and NIPR. Where pooled ages were obtained using data from multiple sessions, mean 206Pb/238U ages were calculated incorporating errors from standard reproducibility in each session. Samples selected for analyses from the NE Terrane comprise four pelitic to semipelitic paragneisses, two hornblende – biotite gneisses (of probable volcaniclastic origin), one enderbitic orthogneiss and one granitic dyke. Samples from the SE Terrane comprise one granitic orthogneiss, one mylonitized granite and one tonalitic orthogneiss. Localities are shown in Figure 3. Zircon and titanite grains were separated from each sample, mounted in epoxy, polished and coated in high-purity gold using an evaporative coater. Sub-grain ion beam analysis of zircon with complex internal zoning requires careful attention to spot positioning, so backscattered electron (BSE) and cathodoluminescence (CL) imaging was performed with a scanning electron microscope before and after analysis, to identify spot positions overlapping multiple growth zones, grain edges, cracks or damaged zircon. Data from analyses or sessions with analytical problems (such as ion beam instability) are not included in the tables. On Tera – Wasserburg plots, data from spots overlapping multiple growth zones are marked with open error ellipses (68.3% confidence level), and analyses on cracked or damaged zircon or with strongly discordant isotopic ratios are marked with error crosses (1s).
Table 1. SHRIMP U–Pb data for zircon from sample 85020401C Spot
232
Th/ U
% 206Pb* Pbc (ppm)
+%
173 128 67
8.299 5.921 10.836
0.08 0.01 0.09 0.03 0.06 0.10 0.04 0.01 0.07
73 185 111 96 27 213 102 158 111
0.30 0.64 0.24 0.07 0.18 0.17 0.03 0.11 0.04
78 52 115 167 132 125 127 240 104
Th (ppm)
Session A2 1.1 2.1 3.1
1667 879 849
13 75 14
0.01 0.09 0.02
0.03 0.01 0.03
Session A3 4.1 5.1 6.1 7.1 8.1 9.1 10.1 11.1 12.1
865 1681 1079 1106 212 2327 1152 1472 1323
39 11 28 44 64 9 11 16 17
0.05 0.01 0.03 0.04 0.30 0.00 0.01 0.01 0.01
Session N3 13.1 13.2 14.1 15.1 16.1 17.1 18.1 19.1 20.1
1002 430 1189 1425 1772 1609 1236 2239 1281
12 50 8 97 29 21 44 9 16
0.01 0.12 0.01 0.07 0.02 0.01 0.04 0.00 0.01
238
206
Total Pb/ 206 Pb
+%
1.60 1.05 1.05
0.06297 0.07367 0.05905
10.187 7.791 8.325 9.867 6.807 9.369 9.732 8.003 10.279
1.32 1.86 2.37 1.39 1.75 1.59 2.06 2.75 1.04
10.997 7.063 8.912 7.313 11.525 11.066 8.340 8.020 10.560
0.58 0.88 1 0.56 0.79 0.89 1.00 1.50 0.60
238
238
Pb/ U age
1s error
207
Pb/ Pb age
1s error
% Discordant
0.06275 0.07361 0.05880
2.57 733.2 0.43 1006.0 0.74 568.9
11.1 9.7 5.7
699.8 1030.9 559.6
54.6 8.8 16.0
25 2 22
Inner rim Detrital Outer rim
1.33 1.86 2.37 1.39 1.75 1.59 2.06 2.75 1.04
0.05931 0.06683 0.06533 0.05999 0.07039 0.06037 0.05978 0.06713 0.06053
0.64 1.45 1.81 0.75 1.88 1.19 1.38 1.60 1.07
603.2 778.4 730.7 622.1 883.2 653.3 630.3 759.0 598.0
7.7 13.6 16.4 8.2 14.5 9.9 12.3 19.7 5.9
578.6 832.5 785.1 603.0 939.6 617.0 595.8 841.9 622.6
13.9 30.3 38.1 16.2 38.5 25.8 29.9 33.4 23.0
24 6 7 23 6 26 26 10 4
Outer rim Inner rim Inner rim Outer rim Mixed Mixed Outer rim Mixed Outer rim
0.59 0.91 1 0.57 0.79 0.90 1.00 1.50 0.60
0.05800 0.06490 0.06310 0.07040 0.05867 0.05903 0.06494 0.06500 0.06040
2.00 3.60 2.40 0.98 1.30 1.50 1.30 2.10 1.20
559.5 848.5 684.0 825.7 535.5 556.8 729.8 757.1 583.0
3.2 7.2 6.5 4.4 4.1 4.8 7.0 10.5 3.4
528.1 772.0 712.9 940.0 554.7 568.3 772.4 773.6 617.8
44.2 75.3 51.0 20.1 27.8 33.2 27.2 44.4 25.0
26 29 4 14 4 2 6 2 6
Outer rim Detrital crack Mixed Mixed Sector z rim Sector z rim Inner rim Rim Sector z core
U/ Pb*
+%
2.55 0.43 0.64
8.302 5.921 10.839
1.60 1.05 1.05
0.05993 0.06689 0.06606 0.06025 0.07085 0.06102 0.06010 0.06722 0.06113
0.56 1.45 1.77 0.68 1.76 1.18 1.35 1.60 1.01
10.194 7.792 8.332 9.870 6.811 9.376 9.736 8.003 10.287
0.06041 0.07020 0.06510 0.07094 0.06010 0.06038 0.06520 0.06590 0.06070
1.30 1.50 2.10 0.75 0.91 1.20 1.20 2.00 1.00
11.030 7.109 8.934 7.318 11.545 11.084 8.343 8.020 10.564
207
206
207
Pb*/ Pb*
+%
206
206
238
206
Notes
AGE CONSTRAINTS IN EAST ANTARCTICA
Total U/ 206 Pb
U (ppm)
(Continued )
27
28
Table 1. Continued Spot
Th (ppm)
235 440 1733 675 525 1241 967 645 297 937 293 258 1184 689 245 1581
101 181 26 74 78 25 16 59 130 16 27 34 22 65 38 9
232
Th/ U
238
0.44 0.42 0.02 0.11 0.15 0.02 0.02 0.09 0.45 0.02 0.10 0.14 0.02 0.10 0.16 0.01
% 206Pb* Pbc (ppm)
206
0.11 0.00 0.30 0.02 0.06 0.01 0.04 0.09 0.03 0.09 0.02 0.16 0.60 0.07 0.71
33 65 193 100 76 106 77 74 44 75 30 39 100 85 35 136
Total U/ 206 Pb
+%
6.030 5.800 7.720 5.810 5.926 10.080 10.840 7.489 5.867 10.790 8.340 5.720 10.210 7.000 6.057 10.010
2.40 2.70 2.90 2.70 1.10 1.10 1.10 1.20 1.10 1.10 2.10 6.60 2.40 2.60 1.10 1.40
238
Total Pb/ 206 Pb
+%
0.07850 0.06960 0.06740 0.07700 0.07325 0.05937 0.05932 0.06610 0.07373 0.05820 0.06740 0.06960 0.06630 0.07715 0.07170 0.06695
4.20 3.90 5.40 2.80 0.63 0.79 0.49 2.40 0.60 2.00 4.50 5.40 2.60 0.42 2.00 1.10
207
238
U/ Pb*
+%
6.030 5.800 7.740 5.810 5.922 10.090 10.840 7.492 5.872 10.790 8.340 5.720 10.220 7.040 6.061 10.080
2.40 2.70 2.90 2.70 1.10 1.10 1.10 1.20 1.10 1.10 2.10 6.60 2.40 2.60 1.10 1.40
206
207
Pb*/ Pb*
+%
0.07760 0.06960 0.06490 0.07690 0.07371 0.05892 0.05927 0.06570 0.07297 0.05800 0.06670 0.06940 0.06490 0.07224 0.07110 0.06114
4.30 3.90 5.60 2.80 0.68 0.84 0.50 2.40 0.73 2.00 4.60 5.50 2.70 0.89 2.00 1.50
206
206
Pb/ U age
1s error
207
Pb/ Pb age
1s error
% Discordant
988.5 1026.1 783.2 1024.2 1005.8 609.2 568.8 807.7 1013.7 571.5 729.8 1038.2 601.6 855.7 984.4 609.5
22.3 25.6 21.1 25.7 10.1 6.2 5.9 9.5 10.4 5.9 14.2 63.4 13.9 20.9 10.5 8.2
1137.3 916.1 771.5 1118.0 1033.5 564.2 577.2 798.3 1013.2 530.5 827.4 909.6 771.6 992.7 960.4 644.2
85.6 79.4 117.8 55.7 13.8 18.3 10.8 51.0 14.8 43.1 96.7 112.4 57.6 18.1 41.8 33.0
15 211 21 9 3 27 1 21 0 27 13 212 28 16 22 6
238
206
Notes
Detrital Detrital Crack Detrital Detrital Outer rim Outer rim Mixed Detrital Outer rim Detrital Detrital Rim Mixed Detrital Outer rim
All errors are quoted at 1s; Pbc and Pb* indicate the common and radiogenic portions, respectively. Common Pb corrected using measured 204Pb. External spot-to-spot errors from observed scatter in standard are 0.00%, 0.00%, 0.00% and 1.01% (respective sessions, included in the calculation of sample-spot errors). Errors in standard calibration are 0.31%, 0.28%, 0.59% and 0.41% (respective sessions, not included in above errors but required when comparing data from different sessions). % Discordance: positive indicates normal (i.e. Pb–Pb age . U –Pb age); negative indicates reverse discordance.
K. SHIRAISHI ET AL.
Session N4 8.2 9.2 21.1 21.2 22.1 23.1 24.1 24.2 25.1 26.1 26.2 27.1 28.1 29.1 29.2 30.1
U (ppm)
Table 2. SHRIMP U–Pb data for zircon from sample 84022004 Spot
U Th (ppm) (ppm)
232
Th/ U
238
% Pbc
206
206
Pb* Total (ppm) 238U/ 206 Pb
+%
+%
0.06970 0.06600 0.06627 0.06539 0.06541 0.06561 0.06560 0.06591 0.07040 0.07070
238 206
U/ Pb*
+%
Pb*/ Pb*
+%
206
Session A2 1.1 2.1 3.1 4.1 5.1 6.1 7.1 8.1 9.1 10.1
130 508 452 431 292 495 196 416 113 102
56 23 220 177 129 270 70 90 42 44
0.44 0.05 0.50 0.42 0.46 0.56 0.37 0.22 0.39 0.44
0.13 0.02 0.03 0.11 0.03 0.11 0.10 0.14 0.69
14 51 54 50 33 56 23 48 13 11
7.780 8.570 7.220 7.376 7.650 7.658 7.510 7.485 7.690 8.390
1.70 1.70 0.64 0.84 0.82 0.87 3.10 0.68 1.90 2.80
7.780 8.590 7.222 7.378 7.659 7.660 7.510 7.492 7.700 8.440
1.30 1.70 1.10 1.10 1.10 1.20 1.80 1.30 1.80 4.00
Session N2 3.2 4.2 7.2 10.2 10.3 17.1 18.1 18.2 19.1 20.1
433 444 254 267 122 238 315 99 263 527
209 258 69 179 54 37 207 34 216 127
0.50 0.60 0.28 0.69 0.46 0.16 0.68 0.36 0.85 0.25
0.24 0.12 0.23 0.29 0.25 0.08 0.44 0.34 0.09 0.15
49 51 39 38 15 24 47 9 38 57
7.615 1.20 0.06510 0.93 7.520 1.50 0.06799 0.81 5.630 2.20 0.07829 0.94 6.002 1.50 0.07398 1.00 6.820 14.00 0.07600 3.60 8.640 1.30 0.06860 2.80 5.765 1.4 0.07708 0.94 9.040 3.10 0.07260 2.70 5.878 1.30 0.07569 0.88 7.944 1.20 0.06859 1.20
7.633 7.530 5.640 6.020 6.840 8.650 5.790 9.070 5.883 7.956
1.20 1.50 2.20 1.50 14.00 1.30 1.4 3.10 1.30 1.20
1.30 1.70 1.10 1.10 1.10 1.20 1.80 1.30 1.80 4.00
207
Pb/ U age
1s error
779.9 710.3 836.0 819.3 791.1 791.0 805.5 807.6 787.2 721.5 793.6 803.6 1051.7 990.7 880.0 705.1 1027.0 674.4 1012.0 763.3
238
0.07040 3.40 0.06480 1.80 0.06606 0.67 0.06514 0.87 0.06442 1.20 0.06539 0.90 0.06460 3.30 0.06503 0.80 0.06920 2.20 0.06450 4.80 0.06308 0.06701 0.07640 0.07150 0.07390 0.06790 0.07340 0.06980 0.07490 0.06739
206
1.40 1.40 1.50 1.80 4.00 3.10 1.70 3.80 1.50 1.30
207
Pb/ Pb age
1s error
% Discordant
9.6 11.7 8.9 8.6 8.5 8.9 13.9 9.9 13.0 27.4
938.7 767.2 808.3 779.0 755.3 786.9 762.0 775.2 903.8 757.4
70.5 38.5 14.1 18.2 24.4 18.9 70.2 16.9 45.2 101.1
20 8 23 25 25 21 25 24 15 5
Igneous Unz. rim Igneous Igneous Igneous Igneous Ig. rim Igneous Igneous Ig. rim
9.2 11.3 21.7 14.1 114.3 9.0 13.7 19.7 12.0 8.7
711.0 838.2 1104.4 972.4 1039.0 865.7 1024.1 922.7 1066.5 849.9
30.0 28.3 29.1 35.7 81.2 63.4 34.4 78.0 29.3 28.0
210 4 5 22 18 23 0 37 5 11
Igneous Igneous Inh. core Inh. core Ig. rim Unz. rim Inh. core Ig. rim Ig. core Ig. crack
206
Notes
AGE CONSTRAINTS IN EAST ANTARCTICA
Total Pb/ 206 Pb
207
All errors are quoted at 1s; Pbc and Pb* indicate the common and radiogenic portions, respectively. Common Pb corrected using measured 204Pb. External spot-to-spot errors from observed scatter in standard are 1.05% and 1.11% (respective sessions, included in the calculation of sample-spot errors). Errors in standard calibration are 0.40% and 0.44% (respective sessions, not included in above errors but required when comparing data from different sessions). % Discordance: positive indicates normal (i.e. Pb– Pb age . U –Pb age); negative indicates reverse discordance.
29
30
Table 3. SHRIMP U–Pb data for zircon from sample 85011503D Spot
200 327 239 93 201 117 256 407 113 114 97 70 82 79 547 825 241 751 281 113 95 217 175 419 169 478 165
42 65 66 16 39 22 47 134 22 22 19 17 15 15 175 281 33 299 88 21 22 69 34 124 38 121 32
232
Th/ U
238
0.21 0.20 0.28 0.17 0.20 0.19 0.18 0.33 0.19 0.20 0.19 0.23 0.19 0.19 0.32 0.34 0.14 0.40 0.31 0.19 0.23 0.32 0.19 0.30 0.23 0.25 0.19
% Pbc
206
0.15 0.01 0.10 0.19 0.02 0.32 1.00 0.25 0.06 0.04 0.09 0.24 0.23 0.10 0.07 0.22 0.03 0.09
206
Pb* (ppm)
22 49 31 13 27 11 26 67 10 11 9 7 7 7 83 127 37 111 44 10 12 27 24 62 27 54 24
Total U/ 206 Pb
238
+%
Total Pb/ 206 Pb
207
+%
238
U/ Pb*
+%
8.974 6.542 7.696 7.003 7.190 10.111 9.493 6.050 10.523 9.858 10.422 9.534 10.502 10.675 6.643 6.558 6.263 6.969 6.398 10.719 7.627 8.118 6.960 6.788 6.157 8.682 6.666
2.66 2.41 4.34 2.85 2.62 2.60 2.54 2.34 2.85 2.90 2.81 3.50 3.09 2.69 2.40 2.32 2.44 2.40 3.83 2.60 3.12 2.45 2.44 2.34 2.73 2.40 2.66
206
207
Pb*/ Pb*
+%
0.06551 0.07118 0.07312 0.07143 0.06926 0.06150 0.06140 0.07079 0.06178 0.05818 0.06191 0.05832 0.06057 0.06053 0.07091 0.07025 0.07214 0.06935 0.07226 0.05991 0.06689 0.06833 0.06546 0.07170 0.06961 0.06717 0.06725
1.54 1.24 7.60 2.21 2.06 1.71 1.58 0.86 1.44 2.94 1.53 6.60 3.42 2.89 0.62 0.73 0.89 0.52 5.31 2.65 2.87 1.93 1.16 0.78 1.72 1.15 1.44
206
206
Pb/ U age
1s error
207
Pb/ Pb age
1s error
% Discordant
Notes
681.1 916.9 787.6 860.5 839.4 608.0 645.6 986.1 585.2 622.8 590.7 643.0 586.4 577.2 903.9 914.9 954.9 864.4 936.1 575.0 794.2 748.9 865.4 885.9 970.3 702.8 901.0
17.2 20.6 32.3 23.0 20.7 15.1 15.6 21.4 16.0 17.2 15.9 21.4 17.3 14.9 20.3 19.8 21.7 19.5 33.5 14.3 23.4 17.4 19.8 19.4 24.6 16.0 22.4
790.8 962.5 1017.4 969.6 906.6 656.7 653.3 951.4 666.5 536.6 670.9 541.9 624.2 622.6 954.8 935.6 989.9 909.2 993.4 600.4 834.4 878.6 789.1 977.6 916.7 843.1 845.6
32.8 25.6 162.1 45.8 43.2 37.2 34.2 17.7 31.2 65.5 33.2 151.4 75.6 63.8 12.8 14.9 18.3 10.7 111.9 58.5 60.9 40.6 24.6 16.1 36.0 24.1 30.2
14 5 23 11 7 7 1 24 12 216 12 219 6 7 5 2 4 5 6 4 5 15 210 9 26 17 27
Unzoned Igneous Ig crack Ig crack Ig core Simple Mix Igneous Simple zoned Simple zoned Rim Rim Rim Simple zoned Igneous Igneous Igneous Ig core Igneous Rim Mixed Ig crack Mixed Igneous Igneous Mixed Igneous
238
206
All errors are quoted at 1s; Pbc and Pb* indicate the common and radiogenic portions, respectively. Common Pb corrected using measured 204Pb. % Discordance: positive indicates normal (i.e. Pb–Pb age . U–Pb age); negative indicates reverse discordance.
K. SHIRAISHI ET AL.
Session A1 1.1 2.1 2.2 3.1 3.2 4.1 5.1 5.2 6.1 7.1 8.1 9.1 10.1 11.1 12.1 13.1 14.1 14.2 15.1 15.2 16.1 17.1 18.1 18.2 19.1 19.2 20.1
U Th (ppm) (ppm)
Table 4. SHRIMP U–Pb data for zircon from sample 9032401A Spot
430 796 363 806 392 651 242 382 627 581 596 1113 234 347 95 520 543 117 185 469 720 79 267 344 213
59 31 37 27 101 52 112 208 216 242 45 10 81 10 36 39 0 48 72 339 393 46 35 123 121
232
Th/ U
238
0.14 0.04 0.11 0.03 0.27 0.08 0.48 0.56 0.36 0.43 0.08 0.01 0.36 0.03 0.39 0.08 0.00 0.42 0.40 0.75 0.56 0.60 0.14 0.37 0.59
206 Pb* Total % Pbc (ppm) 238U/ 206 Pb
+%
206
0.70 0.16 0.56 0.51 0.67 0.44 0.19 0.50 0.33 0.30 0.58 0.09 0.86 0.63 1.53 0.52 0.33 0.27 0.12 0.27 0.14 1.49 0.73 0.26 0.42
38 71 33 73 35 57 22 62 91 95 53 193 21 30 13 47 39 67 65 79 118 11 26 45 39
9.610 1.10 9.645 1.00 9.570 1.20 9.509 10.00 9.580 1.20 9.740 1.00 9.490 1.30 5.260 2.30 5.897 1.30 5.260 2.40 9.600 1.10 4.965 1.30 9.630 1.40 10.000 1.20 6.210 1.60 9.530 1.10 11.930 1.10 1.514 1.40 2.464 1.30 5.105 1.90 5.254 0.98 6.010 1.70 8.780 1.20 6.549 1.10 4.736 1.90
Total Pb/ 206 Pb
207
0.06636 0.06400 0.06595 0.06215 0.06440 0.06407 0.06530 0.08076 0.07837 0.07835 0.06516 0.08009 0.06700 0.06490 0.08290 0.06350 0.06034 0.23750 0.22530 0.07722 0.07632 0.08900 0.06727 0.07547 0.08270
+%
238 206
U/ Pb*
+%
1.40 9.680 1.10 0.94 9.661 1.00 1.40 9.620 1.20 1.20 9.558 1.00 2.50 9.640 1.20 1.10 9.790 1.00 1.70 9.500 1.30 0.92 5.290 2.30 0.79 5.917 1.30 0.76 5.280 2.40 1.30 9.660 1.10 0.52 4.969 1.30 1.70 9.710 1.40 1.60 10.070 1.30 1.90 6.310 1.70 1.10 9.580 1.10 1.30 11.970 1.10 1.00 1.518 1.40 1.10 2.467 1.30 0.89 5.118 1.90 0.66 5.261 0.98 2.00 6.100 1.80 1.50 8.840 1.20 1.10 6.566 1.10 1.50 4.756 1.90
207
Pb*/ Pb*
+%
0.06060 0.06265 0.06130 0.05800 0.05890 0.06050 0.06380 0.07660 0.07560 0.07580 0.06040 0.07931 0.05990 0.05970 0.07020 0.05920 0.05760 0.23510 0.22430 0.07490 0.07518 0.07660 0.06130 0.07330 0.07910
3.40 1.30 3.20 2.10 5.50 2.20 1.80 1.60 1.40 1.40 2.40 0.60 6.00 3.70 6.30 2.90 2.70 1.10 1.10 1.40 0.98 7.60 3.40 1.70 2.30
206
206
Pb/ U age
1s error
207
633.6 634.9 637.3 641.5 636.3 627.2 644.9 1116.0 1006.6 1118.0 635.0 1181.9 631.6 610.5 948.5 640.0 517.3 3261.6 2193.6 1150.4 1121.8 979.2 690.8 913.9 1230.3
6.9 6.1 7.2 6.1 7.6 6.2 7.9 23.3 12.4 24.9 6.4 13.9 8.6 7.3 14.9 6.6 5.4 35.5 24.5 19.7 10.1 16.8 8.1 9.5 21.3
624.6 696.5 650.5 528.5 563.6 621.6 733.5 1109.9 1084.4 1090.5 619.5 1180.0 601.3 593.6 933.4 575.5 515.6 3086.9 3012.1 1067.1 1073.3 1111.7 648.2 1023.3 1175.6
238
Pb/ Pb age
206
1s % error Discordant
73.2 26.8 69.4 46.5 120.2 46.8 39.2 32.4 27.4 27.1 50.7 11.9 128.9 79.2 129.1 62.2 60.2 17.9 18.4 28.1 19.7 151.2 72.5 34.4 45.3
21 10 2 218 211 21 14 21 8 22 22 0 25 23 22 210 0 25 37 27 24 14 26 12 24
Notes
Unzoned Unzoned rim Unzoned rim Core Rim Rim Rim Inherited core Igneous rim Igneous core Sector z rim Core Zoned rim Mixed Ig. crack Core Rim Igneous core Igneous rim Igneous Igneous Ig crack Mixed Igneous core Igneous core
AGE CONSTRAINTS IN EAST ANTARCTICA
Session A4 1.1 2.1 3.1 4.1 4.2 5.1 5.2 6.1 6.2 7.1 7.2 8.1 8.2 10.1 11.1 12.1 12.2 14.1 14.2 15.1 16.1 17.1 18.1 19.1 20.1
U Th (ppm) (ppm)
All errors are quoted at 1s; Pbc and Pb* indicate the common and radiogenic portions, respectively. Common Pb corrected using measured 204Pb. External spot-to-spot error from observed scatter in standard is 0.83% (included in the calculation of sample-spot errors). Error in standard calibration is 0.33% (not included in above errors but required when comparing data from different sessions). % Discordance: positive indicates normal (i.e. Pb–Pb age . U –Pb age); negative indicates reverse discordance.
31
32
Table 5. SHRIMP U–Pb data for zircon from sample 90102801A Spot
U Th (ppm) (ppm)
232
Th/ U
238
206 % Pb* Pbc (ppm)
206
+% Total U/206Pb
238
Total +% Pb/206Pb
207
238
U/ +% Pb*
206
207
Pb*/ Pb*
+%
206
206
1s Pb/ U age error
238
207 206
1s % Pb/ Pb age error Discordant
Notes
190 30 802 72 82 115 114
141 10 78 34 26 56 40
0.74 0.33 0.10 0.47 0.32 0.49 0.35
1.39 1.85 0.14 1.21 1.78 0.83 1.49
32 5 103 12 12 17 20
5.214 5.204 6.011 5.191 5.393 5.585 4.584
2.38 3.67 1.42 2.91 2.47 3.20 5.07
0.08152 0.09703 0.07863 0.08794 0.08143 0.08428 0.08977
2.34 2.82 0.78 2.41 1.60 2.38 4.39
5.288 5.302 6.020 5.255 5.491 5.632 4.653
2.39 3.71 1.42 2.96 2.49 3.22 5.23
0.06968 4.43 0.08140 6.41 0.07746 0.90 0.07773 7.06 0.06613 5.38 0.07724 4.23 0.07720 15.84
1116.6 1113.8 990.7 1123.1 1078.5 1053.6 1255.0
24.6 38.0 13.0 30.6 24.8 31.3 59.8
918.8 1231.2 1133.1 1140.1 810.5 1127.4 1126.3
94.0 131.3 18.2 147.3 116.8 86.7 353.0
222 10 13 2 233 7 211
Igneous Igneous Rim Igneous Igneous Igneous Igneous
Session N1 5b.2 8.1 9.1 9.2 10.1 11.1 12.1 12.2 13.1 13.1b 13.2 14.1 15.1 15.2
318 157 566 161 326 433 562 236 441 346 148 528 2436 144
73 3 57 75 245 76 50 139 96 59 72 247 165 80
0.24 0.02 0.10 0.49 0.78 0.18 0.09 0.61 0.22 0.18 0.50 0.48 0.07 0.57
0.36 0.49 0.14 0.66 0.45 0.31 0.05 0.31 0.58 0.43 0.95 0.26 0.12 1.39
54 25 87 22 51 72 93 39 53 39 26 91 335 24
5.090 5.430 5.613 6.321 5.520 5.148 5.214 5.239 7.184 7.681 4.930 4.988 6.251 5.096
1.20 1.30 1.30 1.40 1.20 1.10 1.20 1.50 1.20 1.20 1.30 1.10 1.20 1.30
0.07914 0.07904 0.07745 0.07775 0.08079 0.07862 0.07768 0.08179 0.07790 0.07613 0.08669 0.08076 0.07632 0.08880
0.76 1.10 0.55 1.20 0.72 0.64 0.82 0.86 0.75 1.10 1.00 0.53 0.31 1.70
5.109 5.457 5.621 6.363 5.545 5.163 5.216 5.256 7.226 7.714 4.978 5.001 6.258 5.168
1.20 1.30 1.30 1.40 1.20 1.10 1.20 1.50 1.20 1.20 1.30 1.10 1.20 1.40
0.07610 0.07490 0.07627 0.07220 0.07700 0.07603 0.07728 0.07920 0.07310 0.07260 0.07860 0.07857 0.07529 0.07710
1152.3 1084.6 1055.5 941.0 1068.9 1141.2 1130.6 1122.8 835.5 785.8 1180.1 1175.2 955.6 1140.3
12.9 13.1 12.6 12.1 11.5 11.9 12.2 15.6 9.2 9.1 14.3 12.0 10.6 14.1
1096.5 1065.2 1102.1 992.5 1122.1 1095.8 1128.4 1177.2 1017.5 1002.7 1163.0 1161.4 1076.4 1123.1
40.2 46.1 16.2 57.8 26.3 18.9 19.8 30.1 30.4 46.1 44.0 15.3 8.5 73.8
25 22 4 5 5 24 0 5 22 28 21 21 13 22
Ig. rim Rim Rim Ig. core Ig. rim Rim Rim Ig. core Rim Rim Ig. core Ig. rim Rim Ig. Core
2.00 2.30 0.81 2.80 1.30 0.95 0.99 1.50 1.50 2.30 2.20 0.77 0.42 3.70
All errors are quoted at 1s; Pbc and Pb* indicate the common and radiogenic portions, respectively. Common Pb corrected using measured 204Pb. External spot-to-spot errors from observed scatter in standard are 0.83% and 1.04% (respective sessions, included in the calculation of sample-spot errors). Errors in standard calibration are 0.33% and 0.44% (respective sessions, not included in above errors but required when comparing data from different sessions). % Discordance: positive indicates normal (i.e. Pb– Pb age . U –Pb age); negative indicates reverse discordance.
K. SHIRAISHI ET AL.
Session A4 1.1 2.1 3.2 4.1 4.2 5.1 6.1
29 56 5 14 4 6 9 4 19 5 Session A7 1.1 191 2.1 71 3.1 10 4.1 22 5.1 17 6.1 10 7.1 13 8.1 9 9.1 116 10.1 12
All errors are quoted at 1s; Pbc and Pb* indicate the common and radiogenic portions, respectively. Common Pb corrected assuming 206Pb/238U – 207Pb/235U concordance. External spot-to-spot error from scatter in standard is 0.93% (included in the calculation of sample-spot errors). Error in standard calibration is 0.38% (not included in errors but required when comparing data from different sessions).
Crack
Crack
6.4 8.0 19.1 14.0 13.9 17.6 15.4 18.5 6.7 15.6 511.8 520.8 443.7 538.2 523.3 461.8 511.0 485.7 520.5 495.0 1.1 1.6 3.8 5.4 2.2 2.7 2.3 3.0 1.4 2.5 0.0658 0.0883 0.2033 0.1680 0.1535 0.1903 0.1726 0.1898 0.0727 0.1774 1.3 1.6 4.2 2.3 2.7 3.8 3.0 3.8 1.3 3.1 11.98 11.44 11.48 9.94 10.43 11.24 10.41 10.69 11.67 10.67 13.7 5.3 0.7 1.9 1.4 0.7 1.1 0.7 8.5 1.0 1.01 3.75 18.21 13.48 11.76 16.54 14.16 16.38 1.83 14.81 0.16 0.81 0.52 0.69 0.24 0.65 0.69 0.43 0.17 0.39
Pb* (ppm)
206
% Pbc 206
Th/238U 232
Th (ppm) U (ppm) Spot
Table 6. SHRIMP U– Pb data for titanite from sample 90102801A
238
Total U/206Pb
+%
207
Total Pb/206Pb
+%
206
Pb*/238U age
1s error
Notes
AGE CONSTRAINTS IN EAST ANTARCTICA
33
NE Terrane Sample 85020401C. This is a migmatitic sillimanite – garnet – biotite paragneiss from the northernmost peak of Perlebandet. The gneiss is intercalated with marble and biotite gneiss showing strong deformation with shear fractures. Garnet porphyroblasts are cracked and filled with secondary muscovite. Monazite, apatite and opaque minerals are also found. Asami et al. (2005) dated monazites from this sample by CHIME (see Discussion). Zircon grains have diverse morphologies and complex internal structures (Fig. 4a). The majority are elongate and rounded, with prismatic and oscillatory-zoned cores enclosed in thin rims of low-CL zircon. Fewer grains are ovoid or equant, with broader low-CL rims that are unzoned or have round, concentric growth zones. Several grains have additional thin outer rims of slightly higher CL, with irregular growth and sector zones, that can be distinguished from inner rims of low-CL zircon (Fig. 4a). The structural complexity required many analyses, and a total of 37 analyses from 30 grains (Table 1) were obtained. From these, seven analyses of spots that covered a mixture of zircon cores and rims (open ellipses, Fig. 5a), and analyses 13.2 and 21.1 on cracked zircon (crosses, Fig. 5a), were excluded from interpretation. Prismatically zoned cores have U contents between 230 and 880 ppm and Th/U ratios between 0.1 and 0.5. Excluding discordant analysis 26.2, seven data from seven cores define a concordia age of 1009 +13 Ma (MSWD ¼ 1.4, probability of concordance ¼ 0.2). The age is of detrital zircon from an igneous source, probably of local derivation (see Discussion). Zircon rims are characterized by high U (850 – 2320 ppm) and very low Th/U (,0.05). Concordant rim ages (from multiple sessions) fall into three discrete populations (Fig. 5a), with 206Pb/238U mean ages of 736 +13 Ma (five data, MSWD ¼ 0.6), 609 +11 Ma (six data, MSWD ¼ 1.3) and 565 + 7 Ma (five data, MSWD ¼ 0.7). Zircon textures or compositions in each age group cannot be distinguished, but are characteristic of zircon growth during granulite-grade metamorphism. In this sample alone, it is uncertain if each age group represents a discrete metamorphic event. It is also not obvious whether the c. 750 Ma age group represents in situ metamorphic growth on detrital (c. 1009 Ma) zircon grains in the metasedimentary host, or if the c. 750 Ma rims are also detrital. Results from other samples (below) help clarify the interpretation of these rim ages. Sample 84022004. This is a foliated medium-grained garnet–biotite paragneiss from Utnibba Nunatak.
34
Table 7. SHRIMP U–Pb data for zircon from sample 90112102A Spot
232
Th/ U
238
206 % Pb* Pbc (ppm)
206
1920 212 177 1176 179 258 390 1277 852 230 343 357 1224 1063 176 385
21 17 11 50 129 5 5 18 36 9 8 6 13 11 0 7
0.01 0.08 0.07 0.04 0.75 0.02 0.01 0.01 0.04 0.04 0.02 0.02 0.01 0.01 0.00 0.02
0.03 0.30 1.18 3.33 0.72 0.09 0.40 2.48 0.03 0.01 0.08 0.60 0.03 0.26 0.01
159 18 15 104 21 21 33 108 70 20 29 30 137 90 15 31
1057 373 451 636 769 286 496 1309
9 8 8 12 18 6 7 28
0.01 0.02 0.02 0.02 0.02 0.02 0.02 0.02
0.06 0.08 0.85 0.06 0.00 0.05 0.02 1.27
85 28 36 53 65 22 39 131
Total U/ 206 Pb
+%
10.350 10.170 10.090 9.700 7.450 10.370 10.110 10.170 10.530 10.120 10.100 10.410 7.700 10.180 9.890 10.830
1.10 1.20 1.20 1.20 2.90 1.20 1.10 1.30 1.10 1.20 1.20 1.10 1.40 1.10 1.30 1.20
10.660 11.570 10.790 10.340 10.240 11.340 11.040 8.590
2.90 3.00 4.90 1.30 4.20 1.40 4.20 3.00
238
+%
238
U/ Pb*
+%
0.05942 0.40 0.06116 1.20 0.06040 3.50 0.07190 2.70 0.09430 3.00 0.06668 1.10 0.06076 1.50 0.06260 0.59 0.07990 4.30 0.06144 1.10 0.06065 0.97 0.05954 1.00 0.07320 1.70 0.05970 0.56 0.06050 3.00 0.06154 10.00
10.360 10.210 10.090 9.820 7.710 10.440 10.120 10.210 10.800 10.130 10.100 10.420 7.750 10.190 9.910 10.830
0.06130 0.05871 0.06753 0.05975 0.05783 0.05858 0.05868 0.07660
10.670 11.580 10.880 10.340 10.240 11.340 11.040 8.700
Total Pb/ 206 Pb
207
206
1.70 1.30 0.77 1.10 1.30 1.10 1.40 3.50
207
206
207
Pb*/ Pb*
+%
1.10 1.20 1.20 1.20 3.00 1.20 1.10 1.30 1.20 1.20 1.20 1.10 1.40 1.10 1.30 1.20
0.05915 0.05850 0.06050 0.06130 0.06450 0.06020 0.05998 0.05896 0.05750 0.06120 0.06059 0.05882 0.06780 0.05941 0.05820 0.06147
0.47 1.80 3.50 3.80 8.20 2.40 1.60 0.99 8.10 1.90 1.00 1.20 2.10 0.63 3.80 1.20
594.2 602.6 609.2 625.4 786.2 589.6 607.5 602.2 571.0 607.1 608.5 591.0 782.4 603.7 619.5 569.5
6.1 6.9 7.1 7.2 22.0 6.7 6.6 7.2 6.3 6.9 6.7 6.5 10.2 6.3 7.7 6.3
572.8 10.2 546.7 38.5 621.7 74.6 651.5 82.2 756.5 173.6 611.2 52.7 602.8 34.2 565.6 21.6 511.5 177.9 645.7 39.8 624.8 21.5 560.6 27.0 862.3 43.9 582.2 13.7 536.4 84.0 655.8 24.8
24 29 2 4 24 4 21 26 210 6 3 25 10 24 213 15
Rim Sector zoned Sector zoned Mixed, crack Inherited Sector zoned Sector zoned Rim Rim, crack Sector zoned Zoned rim Sector zoned Mixed, crack Rim Sector zoned Zoned rim
2.90 3.00 4.90 1.30 4.20 1.40 4.20 3.00
0.06080 0.05803 0.06060 0.05922 0.05781 0.05820 0.05856 0.06620
1.80 1.50 1.70 1.20 1.30 1.30 1.50 4.30
578.0 534.0 567.0 594.9 601.0 544.7 559.0 702.0
16.0 15.0 27.0 7.2 24.0 7.5 22.0 20.0
631.0 531.0 626.0 575.0 523.0 537.0 551.0 812.0
9 21 10 23 213 21 21 16
Rim Sector z, crack Sector zoned Sector zoned Zoned core Zoned rim Sector zoned Mixed, crack
206
Pb/ 1s U error age
238
Pb/ Pb age
206
1s error
38.0 32.0 37.0 25.0 29.0 28.0 32.0 89.0
% Discordant
Notes
K. SHIRAISHI ET AL.
Session A2 1.1 2.1 3.1 4.1 4.2 5.1 6.1 7.1 8.1 9.1 10.1 11.1 12.1 13.1 14.1 15.1 Session N5 1.2 3.2 5.2 6.2 10.2 10.3 11.2 12.2
U Th (ppm) (ppm)
436 3904 213 357 611 704 289 436 635 132 587 180 343 671 715 320 187 843 963
80 99 72 7 10 11 16 6 9 4 6 19 9 8 9 12 155 99 8
0.19 0.03 0.35 0.02 0.02 0.02 0.06 0.01 0.02 0.03 0.01 0.11 0.03 0.01 0.01 0.04 0.85 0.12 0.01
5.92 0.01 0.36 0.64 0.00 0.32 0.11 0.02 1.59 0.01 0.06 0.06 0.80 1.15 0.16 0.07
32 316 27 29 47 55 24 36 54 8 46 15 29 55 56 24 32 78 83
11.920 10.610 6.880 10.750 11.080 11.050 10.570 10.330 10.110 13.330 10.930 10.530 10.250 10.510 10.930 11.360 4.960 9.240 10.010
1.30 5.20 4.20 1.30 1.30 1.30 1.50 1.50 3.90 2.40 1.30 1.30 3.80 2.20 5.40 1.30 8.50 1.50 2.30
0.11060 0.06052 0.07180 0.06506 0.06020 0.06206 0.06150 0.06220 0.06064 0.07710 0.05972 0.06058 0.05895 0.06037 0.06117 0.06525 0.08270 0.06539 0.06025
1.80 1.30 3.00 1.40 1.80 0.64 2.00 3.40 0.69 1.60 0.67 1.40 1.60 0.62 1.40 1.40 1.60 1.10 1.00
12.670 10.610 6.910 10.820 11.080 11.090 10.570 10.340 10.110 13.540 10.930 10.540 10.240 10.520 10.920 11.450 5.010 9.260 10.020
1.40 5.20 4.20 1.30 1.30 1.30 1.50 1.50 3.90 2.40 1.30 1.40 3.80 2.20 5.40 1.30 8.50 1.50 2.30
0.06290 0.06046 0.06880 0.05990 0.06020 0.05946 0.06210 0.06130 0.06047 0.06430 0.05967 0.06010 0.05910 0.05988 0.06128 0.05880 0.07290 0.06412 0.05965
4.70 489.8 1.30 581.0 3.60 872.0 3.10 569.8 1.80 556.9 0.99 556.7 2.00 583.0 3.50 595.1 0.74 608.0 4.10 459.0 0.80 564.2 2.40 584.4 1.60 600.0 0.71 586.0 1.40 565.0 2.10 539.7 4.80 1172.0 1.40 661.4 1.20 613.0
6.4 706.0 29.0 620.0 34.0 893.0 7.3 599.0 6.9 612.0 6.8 584.0 8.4 676.0 8.4 650.0 23.0 620.0 11.0 753.0 6.9 592.0 7.6 608.0 22.0 571.0 12.0 599.0 29.0 649.0 6.7 558.0 91.0 1011.0 9.2 746.0 13.0 591.0
99.0 28.0 75.0 68.0 39.0 21.0 43.0 76.0 16.0 86.0 17.0 51.0 34.0 15.0 30.0 46.0 97.0 29.0 27.0
44 7 2 5 10 5 16 9 2 64 5 4 25 2 15 3 214 13 24
Mixed, crack Rim Inherited Mixed, crack Sector zoned Inherited Sector zoned Sector zoned Zoned core Mixed, crack Sector zoned Sector zoned Sector zoned Sector zoned Sector zoned Sector z, crack Inh, crack Inherited Zoned core
All errors are quoted at 1s; Pbc and Pb* indicate the common and radiogenic portions, respectively. Common Pb corrected using measured 204Pb. External spot-to-spot errors from observed scatter in standard are 1.05% and 1.21% (respective sessions, included in the calculation of sample-spot errors). Errors in standard calibration are 0.40% and 0.40% (respective sessions, not included in above errors but required when comparing data from different sessions). % Discordance: positive indicates normal (i.e. Pb– Pb age . U –Pb age); negative indicates reverse discordance.
AGE CONSTRAINTS IN EAST ANTARCTICA
16.1 17.1 18.1 18.2 19.1 20.1 21.1 22.1 23.1 24.1 25.1 26.1 26.2 27.1 28.1 29.1 30.1 31.1 32.1
35
36
Table 8. SHRIMP U-Pb data for zircon from sample 90112302A Spot
Th (ppm)
1034 156 317 176 2696 356 163 191 232 94 482 362 222 306 147 180 157
56 19 95 14 188 70 18 15 16 13 222 85 20 66 22 25 32
232
Th/ U
238
0.06 0.13 0.31 0.08 0.07 0.20 0.11 0.08 0.07 0.14 0.48 0.24 0.09 0.22 0.15 0.14 0.21
% Pbc
206
0.27 2.42 0.84 6.16 0.32 1.10 3.87 6.12 2.42 7.62 1.56 2.01 1.39 1.72 2.39 1.72 2.81
206
Pb* (ppm)
95 12 29 13 210 37 12 16 18 8 38 28 17 23 12 13 12
Total U/ 206 Pb
+%
9.390 10.860 9.270 11.630 11.050 8.320 11.650 10.160 11.110 10.660 10.800 10.940 11.070 11.600 10.700 11.630 11.150
1.50 2.10 1.70 2.10 1.40 1.70 2.20 2.10 2.00 3.00 1.70 1.90 2.30 2.00 2.70 2.40 2.80
238
Total Pb/ 206 Pb
+%
0.06640 0.07550 0.06660 0.07490 0.06119 0.07220 0.08110 0.08540 0.07750 0.07860 0.06880 0.06670 0.07510 0.07090 0.08390 0.07640 0.08390
2.00 2.70 1.90 2.80 0.79 1.80 5.00 2.70 4.60 4.90 1.90 2.30 5.70 2.90 3.90 3.70 3.80
207
238
U/ Pb*
+%
9.410 11.130 9.350 12.390 11.090 8.410 12.110 10.830 11.390 11.540 10.970 11.160 11.230 11.800 10.960 11.840 11.470
1.50 2.20 1.80 2.50 1.40 1.80 2.50 2.50 2.20 4.40 1.80 2.00 2.60 2.10 3.10 2.90 3.60
206
207
Pb*/ Pb*
+%
0.06420 0.05570 0.05970 0.02300 0.05855 0.06320 0.04900 0.03400 0.05780
2.50 14.00 6.00 51.00 1.50 5.90 20.00 33.00 12.00
0.05600 0.05020 0.06400 0.05690 0.06500 0.06200 0.06100
7.30 9.20 17.00 9.50 21.00 21.00 30.00
206
206
Pb/ U age
1s error
650.9 554.5 654.8 500.4 556.7 724.2 511.3 569.5 542.6 535.7 562.2 553.0 550.0 524.2 562.9 522.8 538.9
9.1 11.9 11.0 12.1 7.6 12.2 12.3 13.7 11.3 22.7 9.6 10.7 13.6 10.8 16.7 14.5 18.7
238
207
Pb/ Pb age
1s error
% Discordant
Notes
206
748.4 441.9 594.0
53.3 305.6 129.2
15 220 29
550.5 713.5 165.5
33.3 125.4 474.6
21 21 268
520.9
269.8
24
453.4 203.6 738.6 489.2 759.9 690.2 643.6
161.0 212.8 350.3 210.2 438.2 450.4 649.3
219 263 34 27 35 32 19
Inh core Igneous Inh core Igneous Rim Inh core Igneous Igneous Igneous Igneous Inh core Inh core Igneous Igneous Igneous Igneous Igneous
All errors are quoted at 1s; Pbc and Pb* indicate the common and radiogenic portions, respectively. Common Pb corrected using measured 204Pb. External spot-to-spot error from observed scatter in standard is 1.35% (included in the calculation of sample-spot errors). Error in standard calibration is 1.08% (not included in above errors but required when comparing data from different sessions). % Discordance: positive indicates normal (i.e. Pb–Pb age . U –Pb age); negative indicates reverse discordance.
K. SHIRAISHI ET AL.
Session A5 1.1 1.2 2.1 2.2 2.3 3.1 3.2 4.1 4.2 5.1 5.2 6.1 6.2 7.1 7.2 8.1 8.2
U (ppm)
Table 9. SHRIMP U–Pb data for zircon from sample 90112302B Spot
131 619 109 175 630 129 85 436 335 658 590 162 775 1238 476 611 507 116 394 882 149 443 291 207 158 358 323 96 508 715 126 67 4021
56 68 25 77 32 42 25 132 146 264 92 52 276 107 242 311 216 39 102 103 50 189 144 71 60 117 135 25 201 313 52 20 101
232
Th/ U
238
0.44 0.11 0.24 0.46 0.05 0.34 0.31 0.31 0.45 0.41 0.16 0.33 0.37 0.09 0.53 0.52 0.44 0.35 0.27 0.12 0.35 0.44 0.51 0.35 0.40 0.34 0.43 0.27 0.41 0.45 0.43 0.30 0.03
% Pbc
206
0.55 0.20 1.43 0.30 1.36 0.36 1.16 0.11 0.09 0.18 0.09 0.83 0.03 0.07 0.15 0.05 0.16 0.63 0.25 0.13 0.42 0.02 0.36 0.29 0.23 0.51 0.41 0.24 0.18 0.33 2.26 0.04
206
Pb* (ppm)
19 65 7 15 64 17 6 65 45 70 52 23 67 112 65 80 43 10 32 96 16 38 38 17 23 30 25 7 43 60 14 8 450
Total U/ 206 Pb
+%
6.050 8.216 13.100 9.951 8.452 6.406 12.000 5.803 6.347 8.120 9.841 6.089 9.982 9.520 6.322 6.563 10.173 9.710 10.683 7.937 8.070 10.146 6.558 10.620 5.960 10.260 11.280 11.170 10.205 10.286 7.550 7.580 7.679
2.00 0.75 1.40 0.93 0.74 0.93 1.10 0.76 1.10 0.73 0.75 0.88 0.73 0.78 1.60 0.75 0.76 1.80 0.78 0.82 7.30 0.88 0.83 2.00 0.91 0.83 1.10 1.10 0.76 0.76 1.30 1.60 0.75
238
Total Pb/ 206 Pb
+%
0.07519 0.06647 0.07120 0.06112 0.07543 0.07442 0.06820 0.07344 0.07222 0.06837 0.06169 0.07778 0.06148 0.06225 0.07374 0.07088 0.06177 0.06961 0.06085 0.06520 0.06905 0.06127 0.07030 0.06233 0.07390 0.06251 0.06396 0.06450 0.06124 0.06048 0.06941 0.08620 0.06427
1.00 0.53 1.50 1.50 1.20 1.10 1.70 0.52 0.63 0.54 0.67 0.85 0.56 0.73 0.54 0.97 0.71 1.40 0.82 0.49 1.10 0.76 0.73 1.20 0.93 0.87 1.00 2.20 0.81 0.90 1.40 2.60 0.35
207
238
U/ Pb*
+%
6.090 8.232 13.290 9.980 8.569 6.428 12.140 5.809 6.353 8.135 9.850 6.141 9.985 9.527 6.331 6.566 10.190 9.770 10.709 7.948 8.100 10.148 6.554 10.660 5.977 10.284 11.340 11.220 10.229 10.305 7.570 7.760 7.683
2.00 0.75 1.50 0.94 0.76 0.95 1.10 0.76 1.10 0.74 0.75 0.90 0.73 0.78 1.60 0.76 0.76 1.90 0.79 0.82 7.30 0.88 0.83 2.00 0.92 0.84 1.10 1.20 0.77 0.76 1.30 1.70 0.75
206
207
Pb*/ Pb*
+%
0.07060 0.06485 0.05970 0.05870 0.06420 0.07150 0.05880 0.07251 0.07147 0.06686 0.06092 0.07080 0.06123 0.06167 0.07246 0.07045 0.06045 0.06450 0.05884 0.06415 0.06560 0.06112 0.07073 0.05940 0.07150 0.06061 0.05980 0.06120 0.05931 0.05898 0.06670 0.06760 0.06392
2.40 0.79 7.00 2.90 2.50 2.30 5.70 0.75 0.86 0.84 0.92 2.20 0.67 0.80 0.76 1.00 1.00 3.40 1.40 0.71 2.20 0.83 0.93 2.90 1.60 1.60 2.10 5.60 1.30 1.20 2.40 6.50 0.36
206
206
Pb/ 1s U error age
238
980.7 739.0 467.6 615.6 711.5 932.0 510.1 1023.9 942.3 747.3 623.3 972.6 615.3 643.4 945.3 913.8 603.5 628.1 575.5 764.0 750.4 605.8 915.4 578.0 997.2 598.2 544.9 550.4 601.3 597.0 799.7 781.6 788.8
17.8 5.3 6.8 5.5 5.1 8.2 5.6 7.2 10.1 5.2 4.5 8.1 4.3 4.8 13.7 6.4 4.4 11.1 4.3 5.9 51.7 5.1 7.1 11.1 8.5 4.8 5.8 6.1 4.4 4.3 10.1 12.1 5.5
207
Pb/ Pb age
1s error
% Discordant
Notes
944.8 769.4 592.9 555.3 749.0 970.7 558.7 1000.4 971.0 833.5 636.4 952.2 647.4 662.6 998.9 941.5 619.6 756.5 561.3 746.5 793.9 643.3 949.6 580.6 971.2 625.3 595.8 644.5 578.7 566.4 828.8 854.9 738.9
49.1 16.6 150.9 62.9 52.1 47.1 125.0 15.3 17.5 17.4 19.7 44.3 14.4 17.2 15.5 20.8 22.4 70.7 31.2 14.9 45.8 17.8 19.0 63.4 33.0 34.0 46.1 120.7 28.5 25.3 51.1 135.1 7.7
24 4 27 210 5 4 10 22 3 12 2 22 5 3 6 3 3 20 22 22 6 6 4 0 23 5 9 17 24 25 4 9 26
Inh core ig Inner rm Leach rim Sector Inner rim Inh core ig Leach rim Inh core ig Inh core ig Mixed Inh core ig crack Out rim / mix? Outer rim Outer rim Inh core ig Inh core ig mix? Outer rim Inh core ig Outer rim Inner rim Inh core frag Outer rim Inh core sector Sector Inh core ig Outer rim Mixed Leach rim Out? unzoned Outer rim INH core ig crack Inh core ig crack Inner rim
206
37
All errors are quoted at 1s; Pbc and Pb* indicate the common and radiogenic portions, respectively. Common Pb corrected using measured 204Pb. External spot-to-spot error from observed scatter in standard is 0.61% (included in the calculation of sample-spot errors). Error in standard calibration is 0.21% (not included in above errors but required when comparing data from different sessions). % Discordance: positive indicates normal (i.e. Pb–Pb age . U –Pb age); negative indicates reverse discordance.
AGE CONSTRAINTS IN EAST ANTARCTICA
Session A6 1.1 1.2 2.1 2.2 3.1 3.3 4.1 4.2 4.3 4.4 4.5 5.1 5.2 6.2 6.3 7.1 7.2 8.1 8.2 9.1 9.2 10.1 10.2 11.1 12.1 12.2 12.3 13.1 13.2 14.1 14.2 15.1 15.2
U Th (ppm) (ppm)
38
Table 10. SHRIMP U–Pb data for zircon from sample 9091405A Spot
232
Th/ U
% Pbc
Th (ppm)
Session A4 2.a 2.1 2.2 3.1 4.1 4.2 5.1 5.2 6.1 8.2 9.1 10.1 10.2 11.1 11.2
302 162 471 412 441 1239 278 3485 271 1550 278 1512 427 10361 3880
202 48 229 172 229 78 107 207 72 90 92 96 142 1005 1232
0.69 0.30 0.50 0.43 0.54 0.06 0.40 0.06 0.28 0.06 0.34 0.07 0.34 0.10 0.33
0.17 1.42 0.72 1.26 0.46 1.86 0.79 0.46 1.16 3.59 1.24 0.45 0.77 0.02 0.04
Session N1 8.3 13.1 14.1 15.1 16.1 17.1
394 589 806 319 787 1029
131 184 68 95 208 39
0.34 0.32 0.09 0.31 0.27 0.04
0.55 0.29 0.24 1.03 3.69 0.25
238
206
206
Total U/ 206 Pb
+%
50 13 38 33 35 100 40 298 20 124 21 119 33 955 318
5.194 10.700 10.710 10.750 10.910 10.611 6.050 10.059 11.470 10.710 11.310 10.920 10.990 9.325 10.496
31 46 63 25 63 81
10.940 10.930 11.080 10.970 10.810 10.910
Pb* (ppm)
Total Pb/ 206 Pb
+%
10.00 1.20 1.10 1.00 1.00 0.92 1.10 0.88 1.20 0.94 1.30 0.93 1.10 0.86 0.88
0.07787 0.07340 0.06580 0.06673 0.06264 0.07403 0.07758 0.06282 0.06450 0.08880 0.06580 0.06240 0.06570 0.05891 0.05971
1.20 1.10 1.10 1.50 1.10 1.10
0.06420 0.06157 0.06056 0.06172 0.09590 0.06110
238
238
U/ Pb*
+%
0.71 1.50 0.93 1.50 1.10 0.65 1.00 0.77 1.70 1.50 1.70 0.72 2.10 0.26 0.44
5.203 10.850 10.790 10.880 10.960 10.810 6.098 10.105 11.610 11.110 11.450 10.970 11.070 9.326 10.499
1.10 0.87 0.77 1.30 6.20 0.83
11.000 10.960 11.110 11.090 11.230 10.940
207
206
207
Pb/ U age
1s error
207
Pb/ Pb age
1s error
% Discordant
Notes
0.94 5.80 2.50 3.70 3.00 2.50 2.20 1.10 5.90 5.20 6.20 1.60 4.00 0.28 0.47
1133.3 568.3 571.3 566.7 562.9 570.2 978.9 608.3 532.8 555.5 539.6 562.4 557.3 656.6 586.5
10.4 7.0 6.3 5.7 5.6 5.2 10.3 5.1 6.6 5.4 6.9 5.1 6.1 5.4 4.9
1107.6 668.1 601.6 470.1 564.4 560.4 956.6 569.1 412.1 587.7 439.9 558.0 582.0 559.0 582.3
18.7 123.7 54.8 81.4 66.1 54.3 45.9 24.1 132.8 113.0 137.0 33.9 87.8 6.1 10.2
22 18 5 217 0 22 22 26 223 6 218 21 4 215 21
Inherited Igneous? Ig. core Ig. core Ig. core Rim Inherited Mixed Ig. core Rim Ig. core Rim Ig. core Inherited Igneous?
2.40 2.20 1.40 4.10 9.70 1.10
560.8 562.7 555.7 556.7 549.9 563.8
6.5 6.2 6.0 8.3 6.1 6.0
594.7 575.1 552.5 341.2 808.5 569.7
51.8 48.6 30.8 93.2 203.8 24.2
6 2 21 239 47 1
Ig. core Ig. core Rim Ig. core Ig. core Rim
Pb*/ Pb*
+%
10.00 1.30 1.20 1.10 1.00 0.95 1.10 0.88 1.30 1.00 1.30 0.94 1.10 0.86 0.88
0.07648 0.06180 0.05990 0.05640 0.05890 0.05880 0.07100 0.05906 0.05500 0.05960 0.05570 0.05876 0.05940 0.05878 0.05942
1.20 1.10 1.10 1.60 1.20 1.10
0.05980 0.05920 0.05861 0.05330 0.06610 0.05907
206
206
238
206
All errors are quoted at 1s; Pbc and Pb* indicate the common and radiogenic portions, respectively. Common Pb corrected using measured 204Pb. External spot-to-spot errors from observed scatter in standard are 0.83% and 1.04% (respective sessions, included in the calculation of sample-spot errors). Errors in standard calibration are 0.33% and 0.44% (respective sessions, not included in above errors but required when comparing data from different sessions). % Discordance: positive indicates normal (i.e. Pb– Pb age . U –Pb age); negative indicates reverse discordance.
K. SHIRAISHI ET AL.
U (ppm)
99 182 216 246 172 101 147 41 181 315 280 315 304 244 256 161 Session N6 1.1 1.2 2.1 2.2 3.1 3.2 5.1 6.1
All errors are quoted at 1s; Pbc and Pb* indicate the common and radiogenic portions, respectively. Common Pb corrected assuming 206Pb/238U – 207Pb/235U concordance. External spot-to-spot error from scatter in standard is 0.85% (included in the calculation of sample-spot errors). Error in standard calibration is 0.41% (not included in errors but required when comparing data from different sessions).
22.3 9.0 5.9 6.0 12.1 8.9 7.6 7.7 561.1 594.0 571.1 570.4 552.1 585.3 560.6 559.5 9.74 6.38 0.51 0.50 7.87 2.58 4.03 0.99 0.2492 0.1286 0.0788 0.0955 0.1653 0.2089 0.1499 0.2860 1.13 1.10 1.06 1.07 1.27 1.28 1.10 1.13 8.430 9.488 10.534 10.327 9.719 8.598 9.778 7.957 18.4 28.5 22.8 26.2 26.9 24.4 22.5 17.4 0.57 0.60 0.80 0.81 0.58 0.43 0.60 0.27
23.34 8.43 2.42 4.47 13.09 18.29 11.16 27.86
206
Pbc 206
% Th/238U 232
Th (ppm) U (ppm) Spot
Table 11. SHRIMP U– Pb data for titanite from sample 9091405A
Pb* (ppm)
Total 238U/206Pb
+%
Total
207
Pb/206Pb
+%
206
Pb*/238U age
1s error
AGE CONSTRAINTS IN EAST ANTARCTICA
39
Other minerals are abundant plagioclase, quartz and minor K-feldspar, occasionally with myrmekites on grain boundaries. Monazite, apatite and opaque minerals are also found. Monazites from this sample were dated by CHIME and yield 542 +12 Ma for rims of zoned grains and 517 +14 Ma for chronologically homogeneous grains (Asami et al. 2005). Zircon grains have a variety of rounded and anhedral morphologies. Both elongate and squat grain types are mostly oscillatory zoned, with elongate grains having lower CL. Cores with oscillatory zoning are truncated at rounded boundaries by oscillatory-zoned rims. A few grains have minor overgrowths of unzoned zircon, on the ends of prismatic cores and rims. U contents of oscillatory-zoned zircon vary between 50 and 530 ppm, with Th/U ratios between 0.2 and 0.9. Analysis 2.1, which has a U content of 508 ppm and a Th/U ratio of 0.05, was performed on a low-CL band between prismatic zircon and an overgrowth. Rounded cores have consistently higher Pb –Pb spot ages than rims and core-free grains (Fig. 4b, Table 2). Most of the core age data are discordant, but four concordant analyses define a concordia age of 1014 + 15 Ma (MSWD ¼ 1.4, probability of concordance ¼ 0.4, Fig. 5b). From oscillatory-zoned grains and rims, five concordant analyses define a concordia age of 790.8 + 9.8 Ma (MSWD ¼ 1.2, probability of concordance ¼ 0.09). The concordia ages are interpreted as timing the growth of igneous zircon at c. 791 Ma, from a magma that inherited zircon from source or country rocks, which grew magmatically at c. 1014 Ma. As the host rock represents metamorphosed sediment, the igneous zircon is detrital, and it is worth noting that all c. 1014 Ma zircon is inherited in c. 791 Ma zircon, so that analysed detrital grains may have derived from the weathering of a single igneous source. Overgrowths of unzoned zircon were not dated, but may be metamorphic. Sample 85011503D. This is an enderbitic orthogneiss from the northwestern part of Brattnepene and consists of plagioclase, quartz, orthopyroxene, brown hornblende, biotite, clinopyroxene and a trace amount of garnet. On the basis of various geothermobarometries, the peak metamorphic conditions were estimated to be around 800 8C and 7– 8.5 kbar (Shiraishi & Kojima 1987). The sample was used for Sm–Nd and Rb–Sr isotope analyses as one of four whole-rock samples (Shiraishi & Kagami 1992). The isochrons yielded 978 + 52 Ma (initial ratio 0.70426) and 961 +101 Ma (initial ratio 0.51163) for the Rb– Sr and Sm –Nd systems, respectively. A Sm –Nd internal mineral isochron yields an age of 624 +18 Ma, with an initial ratio of 0.51193.
Spot
232
Th/ U
% Pbc
Th (ppm)
Session A3 1.1 2.1 3.1 4.1 5.1 6.1 7.1 8.1 9.1 10.1 11.1 12.1 13.1 14.1 16.2 17.1
1398 1744 1550 2269 1917 1099 1774 2265 1054 1988 2796 407 1157 2722 1584 2336
40 177 272 29 108 111 99 94 232 129 156 35 158 74 71 189
0.03 0.10 0.18 0.01 0.06 0.10 0.06 0.04 0.22 0.07 0.06 0.09 0.14 0.03 0.05 0.08
0.24 0.05 0.18 0.03 0.04 0.07 0.16 0.02 0.12 0.02 0.06 0.88 0.85 0.03 0.33 0.24
Session N5 1.3 2.3 5.3 6.3 7.2 8.3 10.3 14.4 14.5 16.3 18.1 18.2 19.1 19.2 20.1 21.1 22.1 23.1 23.2 24.1
102 201 107 720 1907 88 2479 180 2744 89 662 164 2538 52 5419 44 78 99 886 75
47 191 72 88 93 45 180 101 39 65 103 79 33 30 77 23 43 59 47 35
0.48 0.98 0.69 0.13 0.05 0.53 0.08 0.58 0.01 0.75 0.16 0.50 0.01 0.60 0.01 0.54 0.56 0.62 0.06 0.48
0.43 0.29 0.58 0.46 0.06 0.08 0.01 0.03 0.00 0.07 20.01 0.33 0.03 20.24 0.00 20.07 20.15 0.07 0.00 0.04
238
206
206
Total U/ 206 Pb
+%
153 200 176 264 216 125 195 258 121 224 316 44 128 308 107 161
10.944 10.682 11.038 10.286 10.725 10.846 11.027 10.576 11.037 10.777 10.705 11.228 11.174 10.605 11.102 10.946
9 18 9 53 154 8 207 16 225 8 52 14 213 5 508 4 6 9 74 6
9.800 9.390 10.450 11.710 10.620 9.670 10.300 9.940 10.470 9.240 10.840 9.780 10.240 9.290 9.160 10.690 10.470 9.170 10.340 10.610
Pb* (ppm)
Total Pb/ 206 Pb
+%
1.31 1.05 1.06 1.07 1.05 1.15 1.43 1.03 1.13 1.22 1.02 1.23 1.33 1.03 1.17 1.09
0.06069 0.05910 0.06030 0.05934 0.05893 0.05898 0.06005 0.05872 0.05926 0.05975 0.05952 0.06505 0.06505 0.05898 0.06358 0.06164
2.90 3.00 1.50 4.50 2.80 1.80 1.20 5.60 4.60 2.70 1.30 1.30 2.40 1.60 4.50 6.00 2.80 1.70 5.20 2.60
0.06370 0.06400 0.06120 0.06512 0.06247 0.06080 0.05963 0.06383 0.05935 0.06340 0.05858 0.06079 0.06032 0.05980 0.06109 0.06030 0.06020 0.06087 0.05916 0.06060
238
238
U/ Pb*
+%
0.63 0.54 0.33 0.39 0.37 0.70 0.37 0.37 0.40 0.44 0.25 0.72 0.48 0.32 0.66 0.67
10.970 10.688 11.058 10.290 10.730 10.853 11.045 10.578 11.051 10.779 10.711 11.328 11.270 10.608 11.139 10.972
1.90 1.90 1.90 1.10 1.30 1.60 0.30 1.30 0.75 2.20 1.40 1.10 0.68 2.20 1.40 2.90 2.20 1.40 1.30 2.40
9.840 9.420 10.520 11.770 10.630 9.680 10.300 9.940 10.470 9.250 10.840 9.810 10.240 9.270 9.160 10.680 10.460 9.180 10.340 10.610
207
207
Pb*/ Pb*
+%
1.31 1.05 1.06 1.07 1.05 1.15 1.43 1.03 1.13 1.22 1.02 1.24 1.33 1.03 1.17 1.09
0.05871 0.05867 0.05882 0.05907 0.05857 0.05842 0.05874 0.05854 0.05828 0.05957 0.05906 0.05785 0.05810 0.05873 0.06090 0.05968
2.90 3.00 1.50 4.50 2.80 1.80 1.20 5.60 4.60 2.70 1.30 1.40 2.40 1.60 4.50 6.00 2.80 1.70 5.20 2.60
0.06020 0.06160 0.05650 0.06141 0.06198 0.06020 0.05956 0.06360 0.05938 0.06280 0.05865 0.05800 0.06009 0.06180 0.06108 0.06090 0.06140 0.06030 0.05920 0.06020
206
206
Pb/ U age
1s error
0.75 0.58 0.63 0.42 0.41 0.74 0.44 0.38 0.48 0.44 0.27 1.97 1.02 0.32 0.99 0.85
562.4 576.6 558.1 597.9 574.4 568.2 558.7 582.3 558.5 571.9 575.4 545.4 548.0 580.7 554.2 562.3
3.00 2.50 4.30 1.40 1.30 1.90 0.31 1.70 0.75 3.20 1.40 2.40 0.69 2.80 1.40 3.10 3.60 1.70 1.30 2.70
624.0 651.0 585.6 526.0 580.0 634.0 597.3 618.0 588.0 662.0 568.7 625.7 601.0 660.3 668.0 577.0 589.0 667.0 595.0 580.0
206
238
207
Pb/ Pb age
1s error
% Discordant
Notes
206
7.0 5.8 5.7 6.1 5.8 6.3 7.7 5.7 6.0 6.7 5.6 6.5 7.0 5.7 6.2 5.9
556.3 554.7 560.5 569.8 551.1 545.7 557.5 550.1 540.4 588.0 569.5 524.1 533.6 556.9 635.6 592.1
16.3 12.5 13.7 9.1 9.0 16.0 9.6 8.3 10.7 9.6 5.8 43.7 22.3 7.1 21.5 18.5
21 24 0 25 24 24 0 26 23 3 21 24 23 24 13 5
Metm rim Metm rim Rim, crack Rim, crack Metm rim Metm rim Rim, crack Metm rim Mixed, crack Metm rim Metm rim Rim, crack Mixed, crack Metm rim Rim, crack Metm rim
17.0 19.0 8.5 23.0 15.0 11.0 7.1 33.0 26.0 17.0 6.9 8.1 14.0 9.9 28.0 33.0 16.0 11.0 29.0 14.0
610.0 661.0 471.0 654.0 673.0 610.0 587.7 729.0 581.0 702.0 554.0 532.0 607.0 666.0 642.0 635.0 652.0 614.0 574.0 611.0
65.0 54.0 95.0 31.0 28.0 41.0 6.7 36.0 16.0 67.0 31.0 52.0 15.0 59.0 31.0 66.0 78.0 38.0 28.0 59.0
22 2 220 24 16 24 22 18 21 6 23 215 1 1 24 10 11 28 23 5
Ig. core Ig. core Ig. core, crack Rim, crack Rim, crack Ig. core Metm. rim Ig. core Metm. rim Ig. core Metm. rim Ig. core, crack Metm. rim Ig. core Metm. rim Ig. core Ig. core Ig. core Metm. rim Metm. rim
All errors are quoted at 1s; Pbc and Pb* indicate the common and radiogenic portions, respectively. Common Pb corrected using measured 204Pb. External spot-to-spot errors from observed scatter in standard are 1.21% and 1.11% (respective sessions, included in the calculation of sample-spot errors). Errors in standard calibration are 0.40% and 0.44% (respective sessions, not included in above errors but required when comparing data from different sessions). % Discordance: positive indicates normal (i.e. Pb– Pb age . U –Pb age); negative indicates reverse discordance.
K. SHIRAISHI ET AL.
U (ppm)
40
Table 12. SHRIMP U–Pb data for zircon from sample 85012817
Table 13. SHRIMP U–Pb data for zircon from sample 9031507 Spot
232
Th (ppm)
Session A6 1.1 1.2 2.1 2.2 3.1 4.1 5.1 5.2 6.1 7.1 7.2 8.1
70 61 68 271 72 54 97 107 10 257 68 9
32 19 28 95 24 20 55 70 3 55 26 2
0.47 0.33 0.42 0.36 0.34 0.37 0.58 0.68 0.29 0.22 0.40 0.24
0.27 0.91 0.56 0.22 0.20 0.16 0.76 0.91 8.53 0.35 0.84 1.05
9.1
315
101
0.33
0.09
238
206
206
Total U/ 206 Pb
+%
9 8 9 36 10 7 13 14 1 34 9 1
6.443 6.623 6.539 6.521 6.346 6.626 6.459 6.373 7.030 6.586 6.392 6.800
40
6.765
Pb* (ppm)
Total Pb/ 206 Pb
+%
1.20 1.30 1.20 0.96 1.20 1.30 1.40 1.20 2.50 0.97 1.20 2.50
0.07290 0.07620 0.07460 0.06996 0.07270 0.07410 0.07460 0.07381 0.09600 0.07013 0.07290 0.09800
0.95
0.07004
238
238
U/ Pb*
+%
1.90 1.50 1.40 0.73 1.40 1.60 1.90 1.20 4.70 0.76 1.40 3.40
6.461 6.684 6.576 6.536 6.359 6.637 6.508 6.431 7.690 6.609 6.446 6.870
1.30 1.30 1.20 0.96 1.20 1.30 1.50 1.20 3.60 0.98 1.30 2.60
0.69
6.770
0.95
207
206
207
Pb/ U age
1s error
207
Pb/ Pb age
1s error
% Discordant
10.8 11.0 10.5 8.2 10.7 11.3 12.9 10.4 26.7 8.3 11.1 21.7
946.0 887.4 924.4 873.4 958.4 1006.1 875.0 813.4
85.8 86.9 63.1 24.9 42.0 61.0 106.0 74.6
2 21 1 25 2 11 25 213
1.40 4.30 9.10
927.7 898.8 912.6 917.8 941.5 904.7 921.4 931.7 788.2 908.2 929.7 875.8
845.0 803.4 1415.7
30.0 90.5 173.1
27 214 62
0.86
888.1
7.9
908.5
17.8
2
Pb*/ Pb*
+%
0.07060 0.06860 0.06990 0.06816 0.07100 0.07270 0.06820 0.06620
4.20 4.20 3.10 1.20 2.10 3.00 5.10 3.60
0.06723 0.06590 0.08950 0.06933
206
206
238
206
Notes
Igneous Igneous Igneous Igneous Igneous Igneous Igneous Igneous Leached rim Rim Igneous Leached patch Rim
AGE CONSTRAINTS IN EAST ANTARCTICA
Th/ U
% Pbc
U (ppm)
All errors are quoted at 1s; Pbc and Pb* indicate the common and radiogenic portions, respectively. Common Pb corrected using measured 204Pb. External spot-to-spot error from observed scatter in standard is 0.84% (included in the calculation of sample-spot errors). Error in standard calibration is 0.23% (not included in above errors but required when comparing data from different sessions). % Discordance: positive indicates normal (i.e. Pb–Pb age . U –Pb age); negative indicates reverse discordance.
41
42
Table 14. SHRIMP U–Pb data for titanite from sample 9031507 Spot
Th (ppm)
96 74 208 47 113 113 93 87 46 68 100 131 77 45
40 35 62 22 29 41 40 46 24 41 49 29 46 22
232
Th/238U
0.43 0.49 0.31 0.48 0.26 0.37 0.44 0.54 0.54 0.63 0.51 0.23 0.62 0.49
%
206
Pbc
1.78 1.64 1.49 1.78 1.64 1.70 1.62 1.76 2.02 1.78 1.41 1.44 1.85 2.17
206
Pb* (ppm)
7.1 5.1 14.0 3.5 8.2 8.1 6.9 6.3 3.3 5.0 7.2 9.7 5.4 3.3
238
Total U/206Pb
+%
11.63 12.47 12.75 11.71 11.88 12.00 11.62 11.78 11.85 11.68 11.87 11.65 12.25 11.82
1.23 1.13 0.99 1.14 1.02 1.02 1.04 1.04 1.13 1.07 1.03 1.01 1.05 1.44
207
Total Pb/206Pb
0.07230 0.07024 0.06875 0.07221 0.07091 0.07121 0.07105 0.07197 0.07396 0.07226 0.06908 0.06957 0.07218 0.07520
+%
1.02 1.22 0.73 1.37 0.89 0.87 0.96 0.99 1.33 1.10 0.93 0.98 1.04 1.34
206
Pb*/238U age
522.8 489.4 479.8 519.4 512.9 507.7 524.0 516.3 512.2 520.4 514.2 523.4 496.8 512.5
1s error
6.3 5.4 4.7 5.8 5.1 5.1 5.3 5.3 5.7 5.5 5.2 5.2 5.1 7.3
Notes
Young Inclusions
Inclusions
All errors are quoted at 1s; Pbc and Pb* indicate the common and radiogenic portions, respectively. Common Pb corrected assuming 206Pb/238U-207Pb/235U concordance. External spot-to-spot error from scatter in standard is 0.93% (included in the calculation of sample-spot errors). Error in standard calibration is 0.38% (not included in errors but required when comparing data from different sessions).
K. SHIRAISHI ET AL.
Session A7 1.1 2.1 3.1 4.1 5.1 6.1 7.1 8.1 9.1 10.1 11.1 12.1 13.1 14.1
U (ppm)
AGE CONSTRAINTS IN EAST ANTARCTICA
43
Fig. 4. Representative images of zircon grains from samples of the NE Terrane. All images are of cathodoluminescence unless otherwise marked (BSE ¼ backscattered electron). Each SHRIMP analysis spot is labelled with grain analysis number, U content (ppm)/Th/U ratio and 204Pb-corrected 206Pb/238U age with 1s error. Ages that are discordant with 207Pb/235U ages outside 68.3% confidence limits are labelled (d).
Zircon grains are squat to elongate, 50–300 mm long, with rounded, anhedral surfaces. Most grains have oscillatory or graduated prismatic growth zones that decrease in CL brightness from core to rim. All prismatically zoned zircon grains have truncated, rounded edges, which are commonly surrounded by thin rims of high-CL, unzoned zircon. These overgrowths are thickest on the terminations of elongate grains, and are similar to discrete, squat grains of zircon that are unzoned or have gradational or sector zoning (Fig. 4c). Prismatically zoned cores and rims contain 90– 830 ppm U and have Th/U ratios between 0.14 and 0.40 (Table 3). High-CL, simple zoned rims and grains have uniform U contents (90 –120 ppm) and Th/U ratios (0.19–0.23). Six concordant data from prismatic cores define a concordia age of 951 + 17 Ma (MSWD ¼ 1.6, probability of concordance ¼ 0.6, Fig. 5c). In contrast, all eight data from high-CL rims and grains define a concordia age of 602 + 15 Ma (MSWD ¼ 1.3, probability of concordance ¼ 0.01). On the basis of zircon texture and composition, the c. 951 Ma age is interpreted as the time of magmatic zircon growth in the igneous protolith, and the c. 602 Ma age as the time of granulite-grade metamorphism.
Sample 9032401A. This is a strongly sheared sillimanite–garnet–biotite paragneiss from the northern part of Austkampane. Other constituent minerals are plagioclase, K-feldspar, quartz, zircon, apatite and opaque minerals. Secondary muscovite is common. Zircon grains are a mixture of squat to equant, clear grains with some pyramidal faceting preserved, and anhedral grains that contain irregular cores with truncated simple and oscillatory zoning, and mid- to high-CL rims of graduated and sector-zoned zircon (Fig. 4d). The latter rims are texturally identical to the clear, equant zircon grains. Irregular cores have U contents between 80 and 720 ppm U and Th/U values between 0.3 and 0.8 (except analysis 8.1, Table 4). Analysis 8.1 is of an irregular core with no CL, and has an unusually high U content (1113 ppm) and low Th/U ratio (0.01). Clear rims and grains have U contents between 230 and 810 ppm and Th/U ratios with both moderate (0.3–0.5) and low (less than 0.1) values. Ages from seven concordant analyses are grouped at c. 3260 Ma (one grain), c. 1200 Ma (two grains) and c. 1100 Ma (four grains), with the latter four analyses defining a 206Pb/238U mean age of 1125 + 18 Ma
44
K. SHIRAISHI ET AL.
Fig. 5. Tera –Wasserburg concordia diagrams of U– Pb SHRIMP data. (a) Zircon from 85020401C; (b) zircon from 84022004; (c) zircon from 85011503D; (d) zircon from 9032401A.
(MSWD ¼ 0.7) (Fig. 5d). All ages are interpreted as those of detrital zircon with an igneous origin, with the exception of analysis 8.1, which may be metamorphic. With the exception of analysis 12.2, which has a concordant age of c. 517 Ma, data from clear rims and grains define a concordia age of 637 + 6 Ma (MSWD ¼ 1.2, probability of concordance ¼ 0.5). The concordia age is interpreted as timing the growth of zircon during granulite-grade metamorphism. Sample 90102801A. This is a strongly deformed and layered hornblende orthogneiss from Isachsenfjella. Deformed syntectonic granite intrudes the gneiss (Fig. 6b). It consists of plagioclase, quartz, hornblende, biotite, titanite and zircon. Zircon grains are squat to elongate, 50–200 mm long, and mostly anhedral with a few faceted surfaces preserved. Internally, all grains have prismatic and oscillatory zoning subparallel to grain edges. In several grains, minor patches and discontinuous overgrowths of mid- to low-CL zircon are present, although in most cases there are no clear boundaries between cores and rims (Fig. 4e). Prismatic, oscillatory-
zoned zircon has U contents between 140 and 530 ppm U and Th/U ratios between 0.2 and 0.8 (Table 5). Overgrowths have similar U contents (150 –570 ppm U) but lower Th/U ratios (0.02– 0.22), except for analysis 15.1 of a low-CL overgrowth with 2436 ppm U and a Th/U ratio of 0.07. Age data are scattered, with no clear distinction between cores and rims (Fig. 7a). Together, five concordant data from cores and rims define a concordia age of 1133 +12 Ma (MSWD ¼ 1.0, probability of concordance ¼ 0.2). This is interpreted as the time of zircon growth in the magmatic protolith. The significance of rim ages is ambiguous. Textures and U/Th contents could be either late magmatic or metamorphic, with the exception of discordant analysis 15.1, which is more typical of metamorphic zircon growth. There is no evidence of zircon growth after c. 1050 Ma. Titanite is present in the orthogneiss as metamorphic grains intergrown with microcline and hornblende, and as thin rims and overgrowths on ilmenite. Grains analysed (Table 6) show little compositional zoning, and contain up to 200 ppm U and 60 ppm Th. Excluding analyses 3.1 and 6.1
AGE CONSTRAINTS IN EAST ANTARCTICA
45
Fig. 6. Photographs of the selected outcrops. (a) Layered hornblende– biotite gneiss (90112302B) intruded by biotite granite dyke (90112302A), Balchenfjella (eastern part of NE Terrane). (b) Foliated hornblende–biotite gneiss (90102801A), Isachsenfjella (eastern part of NE Terrane). (c) Mylonitic granite (9091405A) at Main Shear Zone, Wiederoefjellet (SW Terrane). (d) Massive tonalitic hornblende gneiss (9031507), Mefjell (SW Terrane).
on cracked titanite, data uncorrected for common Pb define a Model 1 linear array data with a lower intercept age of 516 + 9 Ma (MSWD ¼ 1.4) (Fig. 7b). The same eight analyses define an identical 207Pb-corrected 206Pb/238U mean age of 517 + 8 Ma (MSWD ¼ 1.2). Because of the potential for Pb loss from titanite by diffusion at temperatures over 660 8C (Cherniak 1993), the age may have been reset after growth, and therefore represents only a minimum estimate for the timing of the formation of the metamorphic assemblage. Sample 90112102A. This is a 2 m wide layer of garnet –biotite paragneiss in hornblende gneiss from southern Balchenfjella. It consists of garnet, biotite, plagioclase, K-feldspar, antiperthite, zircon and opaque minerals. Myrmekite replacing K-feldspar and secondary muscovite are present. Zircon grains are rounded and mostly equant or squat, with fewer elongate grains up to 150 mm long. The latter mostly contain irregular or rounded cores with oscillatory zoning, surrounded by rims of mid- to high-CL zircon with simple or
sector zoning. The latter rims are similar to the main population of equant grains (Fig. 8a). To compare analytical results between the ANU and NIPR SHRIMP facilities, data were processed separately from two sessions (Table 7, Fig. 7c and d). Only three analyses (4.2, 18.1 and 30.1) are available from irregular cores, which have U contents of about 200 ppm and Th/U ratios above 0.3. Spot 206Pb/238U ages are scattered between c. 1150 and 760 Ma. The remaining analyses have U contents between 130 and 3400 ppm and Th/U ratios below 0.1, and ages that cluster around 600 Ma. Concordant analyses from session A2 (ANU) define a concordia age of 601 + 6 Ma (MSWD ¼ 1.1, probability of concordance ¼ 0.1, Fig. 7c). Analyses from session N5 (NIPR) are more scattered; however, all younger analyses are from spots on cracked zircon, and when excluded the remaining analyses define a concordia age of 593 + 8 Ma (MSWD ¼ 1.3, probability of concordance ¼ 0.5, Fig. 7d). The concordia ages from the two sessions are identical within error, and are interpreted as timing zircon growth during granulite-grade metamorphism. Ages from irregular
46
K. SHIRAISHI ET AL.
Fig. 7. Tera– Wasserburg concordia diagrams of U– Pb SHRIMP data. (a) Zircon from 90102801A; (b) titanite from 90102801A; (c) zircon from 90112102A (session A2); (d) zircon from 90112102A (session N5).
cores are older but scattered, and these cores are interpreted as detrital zircon. Sample 90112302A. This is a biotite granite dyke intruded across intercalated layers of mafic gneiss and migmatitic biotite felsic gneiss (sample 90112302B), in north Balchenfjella (Fig. 6a). Boundaries with the host gneiss are soft and nebulitic, with entrainment of deformed fragments of host gneiss and no evidence of chill margins. These textures suggest emplacement under high-temperature, ductile conditions. The granitoid is composed of biotite, hornblende, plagioclase, quartz, K-feldspar and titanite, with secondary muscovite and carbonates. Zircon grains are squat and euhedral, with prismatic forms only slightly dominant over pyramidal forms and length to width ratios of 2:1 or less. All grains are oscillatory zoned, with several grains showing a decrease in CL brightness from core to rim (Fig. 8b). Many grains have rounded cores with irregular and oscillatory growth zones (Fig. 8c).
Oscillatory-zoned zircon has U contents mostly between 90 and 310 ppm and Th/U ratios between 0.07 and 0.22, excepting analysis 2.3 of a low-CL rim with 2696 ppm U (Table 8, Fig. 8b). Compositions from rounded cores are variable, with U contents between 310 and 1040 ppm and Th/U ratios between 0.06 and 0.5. 206Pb/238U ages from rounded cores are scattered around c. 720 Ma (analysis 3.1), c. 650 Ma (analyses 1.1 and 2.1), and c. 560 Ma (analyses 5.2 and 6.1). Ten concordant data from oscillatory-zoned zircon define a concordia age of 549 +13 Ma (MSWD ¼ 0.9, probability of concordance ¼ 0.7, Fig. 9a). The age is interpreted as timing magmatic zircon growth in the granite, with rounded zircon cores representing xenocrystic zircon incorporated into the magma. Sample 90112302B. This sample is of hornblende– biotite felsic gneiss from north Balchenfjella, and is host to granitic dykes (sample 90112302A, Fig. 6a). The gneiss is intercalated with mafic
AGE CONSTRAINTS IN EAST ANTARCTICA
47
Fig. 8. Representative images of zircon grains from samples of the NE and SW Terranes. All images are of cathodoluminescence unless otherwise marked (BSE ¼ backscattered electron). Each SHRIMP analysis spot is labelled with grain analysis number, U content (ppm)/Th/U ratio and 204Pb-corrected 206Pb/238U age with 1s error. Ages that are discordant with 207Pb/235U ages outside 68.3% confidence limits are labelled.
layers and is strongly deformed. It is composed of hornblende, biotite, plagioclase, quartz and K-feldspar, with minor epidote, titanite and opaque minerals. Zircon grains include rounded, equant to elongate forms 50– 400 mm long. Most contain oscillatory- or sector-zoned cores truncated by rims of low-CL zircon with simple or no zoning; grains of low-CL sector zoned zircon are also present (Fig. 8c). Most grains also have thin discontinuous rims of high-CL, unzoned zircon. Oscillatory-zoned cores have U contents between 110 and 620 ppm and Th/U ratios between 0.1 and 0.6 (Table 9). Low-CL rims and sector-zoned grains have highly variable compositions, with 200 –4000 ppm U and Th/U ratios between 0.03 and 0.5. High-CL rims (three analyses) have U contents between 80 and 110 ppm and Th/U ratios from 0.2 to 0.3. 206 Pb/238U ages are scattered between 460 and 1030 Ma (Fig. 9b). Data from oscillatory-zoned cores fall into groups at c. 1000 Ma and c. 800 Ma. Most of these analyses fall into the older group, with four out of seven analyses defining a concordia age of 983 + 17 Ma (MSWD ¼ 1.6, probability of concordance ¼ 0.8). Another three
analyses from oscillatory-zoned cores define a concordia age of 795 + 17 Ma (MSWD ¼ 0.5, probability of concordance ¼ 0.6). Analyses from low-CL rims and sector-zoned grains mostly cluster around 600 Ma, except for four analyses with 206Pb/238U ages scattered between 710 and 790 Ma. These older analyses come from the inner sides of low-CL rims (Fig. 8c), and have lower Th/U ratios (,0.15) than the remaining rim analyses (0.3–0.5). Although the older low-CL rims cannot be distinguished texturally from the rest, the differences in ages and Th/U ratios suggest that they represent a distinct stage of zircon growth at c. 750 Ma. Of the remaining analyses of low-CL rims and grains, seven out of nine data define a concordia age of 605 + 7 Ma (MSWD ¼ 1.5, probability of concordance ¼ 0.7). Analyses from thin high-CL rims are scattered between 460 and 550 Ma. An age of growth cannot be defined, but it postdates the growth of low-CL zircon at c. 600 Ma, which grew during high-grade metamorphism. Cores with oscillatory zircon are detrital, derived from igneous sources with ages of c. 1000 and 800 Ma. The significance of high-U, low Th/U rims with ages of c. 750 Ma is unclear, but may indicate that metamorphic
48
K. SHIRAISHI ET AL.
Fig. 9. Tera– Wasserburg concordia diagrams of U–Pb SHRIMP data. (a) Zircon from 90112302A; (b) zircon from 90112302B; (c) zircon from 9091405A (session A4); (d) titanite from 9091405A.
lithologies of this age were also eroded and incorporated into the sedimentary protolith.
SW Terrane Sample 9091405A. This mylonitized granite from Vengen has been called Vengen granite by Shiraishi et al. (1992) or Vikinghoegda granite by Li et al. (2003), and was intruded into a large shear zone (Main Shear Zone) between meta-tonalite and gneisses in the SW Terrane (Fig. 6c). This is a fine- to medium-grained granite, consisting of plagioclase, quartz, K-feldspar, biotite, muscovite, titanite, zircon, apatite, epidote and Fe–Ti oxides. The bulk composition falls in the alkali granite field on the Na2O þ K2O v. SiO2 diagram (Li et al. 2003). Zircon grains are euhedral to subhedral, prismatic and 50–200 mm in length. Grains have concentric and oscillatory zoning, with decreasing CL from core to rim (Fig. 8d). There is little evidence of modification that could be attributed to mylonitization. Few grains contain irregular cores with truncated growth zoning, which are considered to be
xenocrystic. Excluding these cores, U contents increase outwards, with cores having 160 and 800 ppm U and Th/U ratios between 0.3 and 0.7, and low-CL rims having 800–4000 ppm U and Th/U ratios below 0.1. To compare analytical results between the ANU and NIPR SHRIMP facilities, data were processed separately from two sessions. For session A4 (ANU), five core and four rim analyses from six grains together define a concordia age of 564 + 5 Ma (MSWD ¼ 0.9, probability of concordance ¼ 0.2, Fig. 9c). From session N1 (NIPR), all six concordant data from six grains define a concordia age of 559 + 7 Ma (MSWD ¼ 1.2, probability of concordance ¼ 0.9). Incorporating errors in standard calibration from each session, an average age of 562 + 7 Ma is suggested as timing zircon growth in the granitic magma. The high-U rims are not a distinct stage of growth, but represent U enrichment in the late stage of magmatic crystallization. Concordant ages of c. 1130 Ma (analysis 2.a) and c. 980 Ma (analysis 5.1) from xenocrystic cores give some indication of crustal material incorporated into the granitic melt.
AGE CONSTRAINTS IN EAST ANTARCTICA
Titanite can be observed in thin-section as pale yellow subhedral to anhedral grains. Grains are dispersed along the mylonitic foliation in aggregates that appear to represent fractured remnants of pre-existing grains. Larger, less fractured grains up to 0.5 mm long preserve rhombic crystal faces, and in contrast to other titanite-bearing samples (90102801A and 9031507) intergrowths of titanite with metamorphic mineral phases are absent. Grains contain 160–320 ppm U and 40–250 ppm Th (Table 11). Data uncorrected for common lead form a roughly linear array above the concordia on the Tera– Wasserburg plot (Fig. 9d). Excluding analysis 1.2, seven data from five grains fall along a Model 1 discordia line with a lower intercept age of 570 + 8 Ma (MSWD ¼ 1.6). The same analyses define a 207Pb-corrected 206Pb/238U mean age of 568 + 11 Ma (MSWD ¼ 1.4). The titanite age overlaps with that obtained from magmatic zircon, and represents magmatic titanite growth, prior to the development of the mylonitic fabric.
49
Sample 85012817. This is a biotite orthogneiss from Vengen. It occurs close to the Main Shear Zone and shows a strong ductile shear fabric. This rock is interpreted as metamorphosed and deformed granite. Constituent minerals are biotite, microcline, quartz and plagioclase, with minor allanite, titanite, secondary chlorite and muscovite. Zircon grains are mostly elongate, 100–400 mm long and subhedral to anhedral. Cores have oscillatory zoning characteristic of magmatic zircon, truncated by unzoned U-rich rims up to 100 mm wide (Fig. 8e). Zircon rims have U contents between 400 and 5500 ppm and Th/U ratios below 0.2, whereas core analyses have U contents mostly below 200 ppm and Th/U ratios between 1.0 and 0.5. (Table 12). Seven concordant data from cores define a concordia age of 653 + 11 Ma (MSWD ¼ 1.4, probability of concordance ¼ 0.9). Concordia ages for zircon rims were calculated separately for the sessions at ANU (Fig. 10a) and NIPR (Fig. 10b). For session A3 (ANU), nine
Fig. 10. Tera –Wasserburg concordia diagrams of U– Pb SHRIMP data. (a) Zircon from 85012817 (session A3); (b) zircon from 85012817 (session N5); (c) zircon from 9031507; (d) titanite from 9031507.
50
K. SHIRAISHI ET AL.
concordant data from rims define a concordia age of 571 + 5 Ma (MSWD ¼ 1.8, probability of concordance ¼ 0.01), whereas for session N5 (NIPR 5), five data loosely define a concordia age of 587 + 10 Ma (MSWD ¼ 1.6, probability of concordance ¼ 0.9). The c. 653 Ma age for zircon cores is interpreted as timing the crystallization of the igneous protolith, with c. 571 Ma rims grown during high-grade metamorphism. Sample 9031507. This is a tonalitic hornblende orthogneiss from Mefjell. The gneiss is considered to be the eastern extension of meta-tonalite in the southwestern area, but less deformed and preserves a massive plutonic fabric (Fig. 6d). It is composed of plagioclase, quartz, hornblende, biotite, epidote, titanite, apatite and zircon. Zircon occurs as a mixture of elongate prismatic grains up to 300 mm long with rounded edges, and irregular fragments. Internal zoning is euhedral, prismatic and oscillatory, with no inherited cores. Annealed fractures occur in several grains, which merge with very thin, discontinuous rims of unzoned zircon (Fig. 8f). Most grains also have extremely thin (less than 5 mm) rims and (rarely) patches of very high-CL zircon. Oscillatory-zoned zircon has ,320 ppm U and Th/U ratios between 0.3 and 0.7, whereas high-CL rims and patches have U and Th contents of 10 ppm or less (Table 13). Nine concordant age data from six grains define a concordia age of 920 + 8 Ma (MSWD ¼ 1.3, probability of concordance ¼ 0.2), interpreted as the time of crystallization of the tonalitic protolith. Concordant age data from three out of four rim analyses are not significantly different from the magmatic age, and are interpreted as minor annealing and recrystallization of the igneous zircon, rather than growth of new zircon during metamorphism. Titanite is an abundant accessory mineral in the orthogneiss, occurring as grains up to 300 mm long intergrown with hornblende and biotite in a metamorphic assemblage. Grains separated show little compositional zoning, and contain up to 200 ppm U and 60 ppm Th (Table 14). Data uncorrected for common lead cluster above the concordia on the Tera–Wasserburg plot (Fig. 10d). Excluding three analyses, carried out on areas of titanite with micro-inclusions of unknown silicates, 11 data from 11 grains define a 207Pb-corrected 206Pb/238U mean age of 517 + 5 Ma (MSWD ¼ 1.0). As with titanite in sample 90102801A, the result provides a minimum age for the growth of metamorphic titanite.
Summary of SHRIMP dating The SHRIMP results are complex, and the interpretation requires an approach that integrates information from textures and compositions of zircon
with structural and compositional relationships in the host lithologies. Zircon and titanite age populations identified in all 11 samples, and the magmatic or metamorphic origin of zircon as determined from grain textures and compositions, are summarized in Figure 11. In the NE Terrane, the majority of samples are from localities of gneisses that are highly deformed and that do not preserve predeformational lithological contacts. Gneiss samples 84022004, 85020401C, 9032401A and 90112102A are garnet-bearing and aluminous, characteristic of a sedimentary origin, and sample 90112302B comes from heterogeneous laminated gneiss interlayered with aluminous and calcareous metasedimentary rocks, and is likely to have a volcaniclastic protolith. Sample 90102801A of hornblende – biotite gneiss is also compositionally layered, contains lenses of calcareous composition, and occurs in a sequence of aluminous gneisses, and therefore is possibly volcaniclastic. Sample 85011503D is interpreted as a metamorphosed igneous enderbite, consistent with the compositionally uniform nature of the lithology in outcrop, which was emplaced at c. 951 Ma. Pre-700 Ma zircon cores and grains are mostly of magmatic origin, and provide a detrital age signature for the terrane, with magmatic activity in three discrete time windows at c. 1130 Ma, c. 1000 Ma and c. 800 Ma (Fig. 11). This simple detrital signature may indicate a local derivation of sedimentary detritus, from synsedimentary volcanism or basement lithologies. Excluding one detrital magmatic zircon with an age of c. 3200 Ma, the lack of pre-1200 Ma ages in the detrital signature indicates a relatively juvenile provenance for the sediments, with no significant contribution from early Proterozoic or Archaean continental crust. Meso-Neoproterozoic sedimentation is also proposed by various researchers from other East Gondwana crustal fragments, such as southern India (Santosh et al. 2006; Collins et al. 2007). Magmatic zircon in the metasedimentary rocks must be detrital in nature, deposited prior to the formation of the gneisses through granulite-grade metamorphism. The presence of c. 800 Ma magmatic zircon in three samples of metasedimentary rocks therefore constrains subsequent metamorphism to a younger age. There is a high productivity of metamorphic zircon (in five samples) within a window between 640 and 600 Ma, with most age populations at c. 600 Ma. This is considered to be the time of zircon growth at or near peak metamorphism and deformation. Evidence for previous metamorphic events from zircon analysis is more ambiguous. Two samples of metasedimentary rocks have high-U, low Th/U overgrowths on
AGE CONSTRAINTS IN EAST ANTARCTICA
51
Fig. 11. Summary of SHRIMP zircon and titanite U– Pb ages from the Sør Rondane Mountains.
magmatic zircon cores with ages of c. 750 Ma, and may be attributed to metamorphism. However, the textural relationships within zircon grains are not clear enough to say whether this zircon represents metamorphism of the host metasedimentary rocks, or of source rocks that provided detritus to sediments deposited after 750 Ma. Because of the rarity of this generation of zircon in the analysed samples in general, and the lack of field evidence for deformation and metamorphism prior to the formation of gneisses at c. 600 Ma, the hypothesis that 750 Ma zircon is detrital is tentatively proposed. However, there is no textural and morphological evidence. Analyses of high-U, low-Th/U zircon that might be attributed to metamorphic growth at c. 1000 Ma or c. 1130 Ma are scant, being restricted to isotopically discordant analyses of zircon rims from sample 90102801A and a xenocrystic zircon core from sample 9032401A. The formation of high-U, low-Th/U rims can also occur in magmatic zircon, where fractional crystallization concentrates U in late-stage magmatic fluids, and this mechanism is invoked to explain euhedral high-U rims in magmatic zircon from samples 90112302A and 9091405A. Consequently, the few U-rich analyses with c. 1000 Ma ages do not represent substantial evidence of high-grade metamorphism at this time.
Zircon growth after c. 600 Ma is also very limited, being restricted to a minor population of c. 560 Ma metamorphic zircon from sample 85020401C, and a few isotopically discordant analyses from thin high-CL rims that are present in many samples. The latter may be attributed to marginal recrystallization and U leaching by hydrothermal fluids after the metamorphic event (Geisler et al. 2003). Magmatic zircon from sample 90112302A dates the intrusion of nebulitic granitic dykes at c. 550 Ma. The relatively undeformed nature of these dykes demonstrates that high-strain deformation had waned by this time. Zircon dating results from the SW Terrane have some important differences from those from the NE Terrane. All three samples derive from metamorphosed and deformed igneous lithologies, with each timing a separate stage of magmatism at c. 920 Ma, c. 650 Ma and c. 560 Ma. The latter age derives from granite that intrudes across a gneissic fabric that is also present in the c. 650 Ma orthogneiss, and therefore constrains the timing of high-strain deformation between 650 and 560 Ma. Metamorphic zircon growth is recognized in sample 85012817 only, and the c. 570 Ma age may coincide with near-peak (amphibolite-grade) metamorphism and deformation in the orthogneiss. If this is the case, it suggests different metamorphic histories for the
52
K. SHIRAISHI ET AL.
SW and NE Terranes, prior to their current juxtaposition by later faults and/or shear zones. The age of mylonitization in association with the Main Shear Zone, which juxtaposes the c. 920 Ma meta-tonalites with amphibolite-grade gneisses in the SW Terrane, is constrained by the c. 560 Ma intrusive age of the sheared granite at Vengen (sample 85012817), and preservation of the intrusive age in titanite from the same sample suggests that this area was not subject to thermal metamorphism after this time. In contrast, titanite ages of c. 517 Ma were obtained from both the metatonalite in the SW Terrane and hornblende– biotite gneiss from the NE Terrane. Both of these localities (Mefjell and Isachsenfjella) lie in the vicinity of large bodies of post-tectonic granite (Li et al. 2003). The c. 517 Ma titanite ages represent either new titanite growth or isotopic resetting through Pb loss, under elevated thermal conditions that may be a result of cooling from an earlier metamorphic peak or of contact metamorphism caused by the emplacement of granitic plutons, as suggested by Asami et al. (1992). In either case, the ages indicate that the NE and SW Terranes share a common history by c. 517 Ma.
Nd model ages Samples and procedures for Nd isotopic analysis Nd model ages have been used by many workers for revealing histories of crustal genesis (e.g. DePaolo 1988; Dickin 1995; Stern 2002; Kagami et al. 2006). In this study we have compiled 180 Nd isotope data from previous studies of central to eastern Dronning Maud Land and Enderby Land, along with 31 new data, mainly from the LHC (Table 15). The new data are derived from samples of paragneiss and basic to intermediate orthogneiss. Quartzofeldspathic gneisses and granites are also included. Some paragneisses have migmatitic textures. Nd isotope analytical procedures follow those of Kagami et al. (1987, 2006). Isotope analyses were performed on a MAT261-type mass spectrometer equipped with five Faraday cups at Niigata University. 143 Nd/144Nd ratios were normalized to 146 Nd/144Nd ¼ 0.7219. 143Nd/144Nd ratios are reported relative to 143Nd/144Nd ¼ 0.511858 for La Jolla or 0.512640 for BCR-1. Sm and Nd concentrations were measured by the isotope dilution method using a 149Sm – 150Nd mixed spike. We estimate an error of 0.1% for the Sm/Nd ratio of each sample based on reproducibility of the data. Depleted mantle Nd model ages (TDM) in this study were calculated using an Excel
spreadsheet provided by Stern (2002), in which the model in Nelson & DePaolo (1984) and DePaolo (1988) was applied. Most of the previous Nd isotope studies (Shiraishi & Kagami 1992; Owada et al. 1994, 2001; Shiraishi et al. 1995, 1997b; Yoshida et al. 1999; Suzuki et al. 2001, 2006; Nishi et al. 2002; Ajishi et al. 2004; Kawano et al. 2005) were performed at the same laboratory as the present study, and 143 Nd/144Nd ratios are reported relative to 143 Nd/144Nd ¼ 0.511858 (La Jolla), with the same isotopic parameters being used. The 147Sm/144Nd ratios in Table 15 vary widely across the region. Higher 147Sm/144Nd ratios are regarded to yield unreliable Nd model ages (e.g. Stern 2002). In this study, data interpretation was restricted to samples with147Sm/144Nd ratios of 0.13 or less, representing 96 analyses out of a total of 180 (Table 15, Fig. 12). Following Stern (2002), analyses with147Sm/144Nd ratios of 0.165 or less represent more than 87% (156 data) of the total dataset.
Regional distribution of Nd model ages A compilation of depleted mantle Nd model ages (TDM), including new analyses and those from previous works recalculated following Stern (2002), provides new insights into crustal residence time in each terrane (Fig. 13). TDM values for the Napier Complex were reported from four areas: Mt. Riiser-Larsen (3.03–3.43 Ga), Tonagh Island (3.10– 3.71 Ga), Mt. Sones (3.43–3.95 Ga) and Fyfe Hills (3.27–3.52 Ga). Tonalitic orthogneiss from Mt. Sones, which has yielded the oldest age in the Napier Complex, also provides the oldest TDM (3.9 Ga) (Black et al. 1986). TDM values for 12 para- and orthogneisses from the Rayner Complex vary widely from 1.27 to 2.28 Ga, with no indications of an Archaean crustal component. In particular, orthogneisses from Sandercock Nunataks, located inland to the south of the Napier Complex, have consistent 1.61 –1.71 Ga ages, postdating the latest pervasive metamorphic and magmatic event at 2.5 Ga in Napier Complex, and suggesting the presence of late Palaeoproterozoic juvenile crust. The only indication of an Archaean component in the RC comes from a single c. 2.6 Ga xenocrystic zircon core in pelitic gneiss, from Mt. Underwood in the Nye Mountains (Shiraishi et al. 1997b). This is significantly older than the c. 1.7 Ga TDM of the same gneiss. TDM values for 37 samples from the LHC and YBC also present a wide range of 0.87 –2.70 Ga, with a major mode at 1.0–1.25 Ga and a small mode at 2.29–2.70 Ga. The latter group derives from both ortho- and paragneisses from the
Table 15. Nd model ages for the rocks from eastern Dronning Maud Land and western Enderby Land, East Antarctica. No.
Sample
Enderbitic gneiss Enderbitic gneiss Enderbitic gneiss Garnet granite Enderbitic gneiss Enderbitic gneiss Enderbitic gneiss Retrograde gneiss Retrograde gneiss Retrograde gneiss Retrograde gneiss Granite Granite Granite Granite Granite Granite Granite Bt qtzfelds gneiss Bt qtzfelds gneiss Bt qtzfelds gneiss Bt qtzfelds gneiss Px–pl gneiss
Locality
Brattnipene Brattnipene Brattnipene Brattnipene Brattnipene Brattnipene Brattnipene Brattnipene Brattnipene Brattnipene Brattnipene Dufek Dufek Dufek Dufek Mejell Mejell Pingvinane Balchen Balchen Balchen Balchen Balchen
Nd Sm (ppm) (ppm)
147
Sm/144Nd
143
Nd/144Nd T(DM) Ga
4.93 3.14 2.73 14.50 3.44 2.37 3.31 1.22 3.77 6.56 4.86 6.91 10.71 9.80 7.58 3.02 18.77 15.65 1.55 2.90 4.81 1.80 3.67
20.03 12.19 12.00 96.40 13.64 12.50 14.60 4.28 14.48 27.69 20.42 44.70 69.11 71.04 61.32 18.54 197.38 84.31 7.48 18.70 14.20 10.90 12.80
0.1488 0.1557 0.1375 0.0909 0.1525 0.1146 0.1371 0.1723 0.1574 0.1432 0.1439 0.0934 0.0937 0.0834 0.0747 0.0985 0.0575 0.1122 0.1253 0.0937 0.2048 0.0998 0.1733
0.512560 0.512620 0.512490 0.512273 0.512540 0.512360 0.512470 0.512720 0.512620 0.512550 0.512530 0.512302 0.512297 0.512278 0.512230 0.512306 0.512101 0.512381 0.512130 0.512180 0.512750 0.512110 0.512770
1.13 1.11 1.11 0.96 1.25 1.05 1.14 1.14 1.07 1.12 0.94 0.95 0.90 0.90 0.98 0.92 1.00 1.57 1.10 1.25
References
Shiraishi & Kagami 1992 Shiraishi & Kagami 1992 Shiraishi & Kagami 1992 This study Shiraishi & Kagami 1992 Shiraishi & Kagami 1992 Shiraishi & Kagami 1992 Shiraishi & Kagami 1992 Shiraishi & Kagami 1992 Shiraishi & Kagami 1992 Shiraishi & Kagami 1992 Arakawa et al. 1994 Arakawa et al. 1994 Arakawa et al. 1994 Arakawa et al. 1994 Arakawa et al. 1994 Arakawa et al. 1994 Arakawa et al. 1994 Grew et al. 1992 Grew et al. 1992 Grew et al. 1992 Grew et al. 1992 Grew et al. 1992
Yamato–Belgica Complex 24 Y80A529 Opx–bt gneiss 25 A79121511 Bt gneiss
Yamato Mts. Belgica Mts.
4.10 1.14
23.36 8.81
0.1061 0.0782
0.512101 0.512214
1.33 0.94
This study This study
Lu¨tzow-Holm Complex 26 80S74 27 80S15 28 80S52 29 80S57 30 80S59
Sinnan Rock Sinnan Rock Sinnan Rock Sinnan Rock Sinnan Rock
2.70 5.40 3.00 2.30 5.20
11.35 25.70 15.70 9.80 26.10
0.1438 0.1270 0.1155 0.1419 0.1204
0.512341 0.512289 0.512229 0.512335 0.512239
1.52 1.32 1.26 1.49 1.31
This study This study This study This study This study
Sil–bt –grt gneiss Bt amphibolite Bt granite Grt–bt gneiss Bt gneiss
53
(Continued )
AGE CONSTRAINTS IN EAST ANTARCTICA
Sør Rondane Mountains 1 85011503B 2 85011503C 3 85011503D 4 85011504B 5 85011602D 6 9022502A 7 9022502B 8 85011601A 9 85011601C 10 85011602A 11 85011602B 12 B9001-2301A 13 2302A 14 2303A 15 2305B 16 2502 17 2405C 18 1406 19 EG88011109 20 EG88011212 21 EG88011319 22 EG88012804 23 EG88012105
Rock type
No.
80S78 73123106 73123103 73123116 74010606 74010701 73123106K 74010105 74010115 74010107 74010113 74010304 No.1 No.2 No.3 No.4 No.5 No.6 K95010804 K950110m2 K950110m3 K950110m4 K950110m6 K950110m7 K950110m8 K950110m9 K950110m10 0101-2 0102A 0102B 0107 1802 95020203 95020205 o3020512 95020301 95020303 o3020616
Rock type Bt gneiss Meta-trondjemite Meta-trondjemite Meta-trondjemite Meta-trondjemite Meta-trondjemite Meta-trondjemite Meta-trondjemite Meta-trondjemite Meta-trondjemite Meta-trondjemite Sil–bt –grt gneiss Bt gneiss Granite Granite Granite Granite Granite Granite Hbl–bt gneiss Hbl–bt gneiss Hbl–bt gneiss Hbl–bt gneiss Hbl–bt gneiss Hbl–bt gneiss Hbl–bt gneiss Hbl–bt gneiss Bt–hbl gneiss Bt–hbl gneiss Qtzfeld gneiss Bt gneiss Grt–bt gneiss Granite Granite Granite Granite Granite Granite
Locality Sinnan Rock Cape Hinode Cape Hinode Cape Hinode Cape Hinode Cape Hinode Cape Hinode Cape Hinode Cape Hinode Cape Hinode Cape Hinode Cape Hinode Kasumi Rock Kasumi Rock Kasumi Rock Kasumi Rock Kasumi Rock Kasumi Rock Oku-iwa Rock Oku-iwa Rock Oku-iwa Rock Oku-iwa Rock Oku-iwa Rock Oku-iwa Rock Oku-iwa Rock Oku-iwa Rock Oku-iwa Rock Nesoya Nesoya Nesoya Nesoya Nesoya East Ongul East Ongul East Ongul East Ongul East Ongul East Ongul
Nd Sm (ppm) (ppm) 3.80 0.72 1.67 1.18 3.60 0.90 1.00 0.23 0.52 2.39 2.51 5.65 3.55 0.27 0.25 0.12 0.23 0.41 22.80 7.37 6.48 4.23 3.94 5.19 8.79 4.33 1.50 10.92 10.05 1.53 3.34 6.24 14.77 12.30 0.95 20.25 11.54 8.36
19.10 4.80 7.86 5.70 14.54 5.84 5.72 2.06 3.30 11.40 11.62 27.98 23.90 1.00 0.95 0.55 0.82 1.18 163.00 26.00 22.90 18.40 23.00 29.30 33.30 18.10 9.39 50.61 45.57 7.01 22.31 35.75 68.83 57.10 5.74 89.61 75.80 36.33
147
Sm/144Nd 0.1203 0.0907 0.1284 0.1251 0.1497 0.0932 0.1057 0.0675 0.0953 0.1267 0.1306 0.1220 0.0897 0.1651 0.1615 0.1332 0.1713 0.2095 0.0845 0.1714 0.1711 0.1390 0.1036 0.1071 0.1596 0.1446 0.0966 0.1304 0.1333 0.1319 0.0905 0.1055 0.1297 0.1302 0.1000 0.1366 0.092 0.1391
143
Nd/144Nd T(DM) Ga
0.512249 0.512173 0.512383 0.512384 0.512510 0.512141 0.512220 0.511985 0.512159 0.512390 0.512427 0.511943 0.512139 0.512303 0.512307 0.512230 0.512258 0.512475 0.511928 0.512680 0.512679 0.512533 0.512378 0.512400 0.512820 0.512553 0.512263 0.512372 0.512376 0.512407 0.512202 0.512315 0.512402 0.512474 0.512286 0.512435 0.512231 0.512438
1.29 1.08 1.18 1.13 1.26 1.14 1.16 1.10 1.14 1.14 1.13 1.82 1.11 2.15 1.53 1.31 1.04 0.92 0.92 0.65 1.08 1.02 1.23 1.26 1.18 1.04 1.03 1.16 1.04 1.02 1.20 1.02 1.23
References This study Shiraishi et al. 1995 Shiraishi et al. 1995 Shiraishi et al. 1995 Shiraishi et al. 1995 Shiraishi et al. 1995 Shiraishi et al. 1995 Shiraishi et al. 1995 Shiraishi et al. 1995 Shiraishi et al. 1995 Shiraishi et al. 1995 Shiraishi et al. 1995 Ajishi et al.2004 Ajishi et al.2004 Ajishi et al.2004 Ajishi et al.2004 Ajishi et al.2004 Ajishi et al.2004 Nishi et al. 2002 Nishi et al. 2002 Nishi et al. 2002 Nishi et al. 2002 Nishi et al. 2002 Nishi et al. 2002 Nishi et al. 2002 Nishi et al. 2002 Nishi et al. 2002 This study This study This study This study This study Kawano et al. 2005 Kawano et al. 2005 Kawano et al. 2005 Kawano et al. 2005 Kawano et al. 2005 Kawano et al. 2005
K. SHIRAISHI ET AL.
31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68
Sample
54
Table 15. Continued
Rayner 105 106 107
o3020608 95020308 o3020504 o3020506 95020305 95020402 o3020308 95020405 o3020601 o3020517 79022008 79022016 79100201 79100202 950205M6 81012802 Y70012905 68031601 Y69101014 Y69101023 A2 A3 A4 A5 2601 2602 2702 C1 C4 C7 20111A Y69020613 Y69020614 Y70020515 RH19B 84011105 Complex MA88021608 78285009 78285010
Granite Granite Granite Granite Granite Granite Granite Granite Leucosome in Hbl gn Leucosome in Hbl gn Bt–opx amphibolite Bt–hbl–opx gneiss Opx granulite Opx–hbl granulite Metamorphic rock Sil–bt –grt gneiss Grt gneiss Noritic charnockite Charnockitic band Granitic gneiss Bt–grt gneiss Bt–grt gneiss Grt–opx –bt granulite Grt–opx –bt gneiss Grt–two –px–bt granulite Grt–qtzfels gneiss Grt–qtzfels gneiss Grt–bt gneiss Px granulite Grt–bt gneiss Sil–bt –grt gneiss Metabasite Grt dioritic gneiss Hbl charnockite Sil–bt –grt gneiss Hbl–bt gneiss
East Ongul East Ongul East Ongul East Ongul East Ongul East Ongul East Ongul East Ongul East Ongul East Ongul East Ongul East Ongul East Ongul East Ongul East Ongul West Ongul Ongul Strait Utholmen Fleynoya Fleynoya Skarvsnes Skarvsnes Skarvsnes Skarvsnes Skarvsnes Skarvsnes Skarvsnes Skarvsnes Skarvsnes Skarvsnes Telen Skallen Skallen Skallevikhalsen Rundvagshetta Innhovde
10.31 16.83 15.51 12.81 15.62 10.40 17.45 2.70 9.40 2.61 0.78 5.33 1.65 3.57 5.77 3.15 5.15 7.69 7.91 10.20 5.57 6.39 14.15 5.56 9.40 4.25 10.20 1.09 14.23 5.24 1.87 7.76 1.43 9.69 3.46 6.13
45.41 79.30 76.95 58.06 62.28 57.20 86.29 9.74 43.94 28.58 3.81 22.16 6.40 13.90 41.85 12.09 21.10 35.70 46.40 41.70 21.22 26.72 56.60 20.59 37.21 16.57 52.42 4.68 65.20 25.80 7.11 29.90 9.38 50.80 21.48 30.37
0.1372 0.1283 0.1218 0.1334 0.1516 0.1099 0.1222 0.1676 0.1293 0.0552 0.1238 0.1454 0.1559 0.1553 0.0833 0.1575 0.1475 0.1302 0.1030 0.1479 0.1587 0.1445 0.1511 0.1631 0.1528 0.1552 0.1176 0.1412 0.1319 0.1228 0.1588 0.1569 0.0921 0.1153 0.0973 0.1220
0.512375 0.512439 0.512383 0.512396 0.512525 0.512233 0.512361 0.512410 0.512359 0.512142 0.512383 0.512548 0.512500 0.512555 0.512295 0.512420 0.512519 0.512377 0.512211 0.512387 0.512481 0.512363 0.512499 0.512502 0.512509 0.512405 0.512246 0.512400 0.512340 0.512290 0.511605 0.512000 0.511220 0.511310 0.511121 0.512224
1.33 1.08 1.10 1.23 1.27 1.19 1.14 1.23 0.87 1.12 1.11 1.42 1.27 0.88 1.68 1.20 1.21 1.15 1.52 1.55 1.49 1.32 1.62 1.33 1.65 1.26 1.33 1.29 1.25 4.22 2.84 2.29 2.70 2.53 1.23
Kawano et al. 2005 Kawano et al. 2005 Kawano et al. 2005 Kawano et al. 2005 Kawano et al. 2005 Kawano et al. 2005 Kawano et al. 2005 Kawano et al. 2005 Kawano et al. 2005 Kawano et al. 2005 Kawano et al. 2005 Kawano et al. 2005 Kawano et al. 2005 Kawano et al. 2005 Kawano et al. 2005 This study Yoshida & Kagami 1995 Yoshida & Kagami 1995 Yoshida & Kagami 1995 Yoshida & Kagami 1995 This study This study This study This study This study This study This study Tanaka et al. 1985 Tanaka et al. 1985 Tanaka et al. 1985 This study Yoshida et al. 1999 Yoshida et al. 1999 Yoshida et al. 1999 This study This study
Pelitic gneiss Paragneiss Pegmatite
Mt. Vechernaya Mt. Underwood Mt. Underwood
5.66 6.43 36.20
17.10 33.40 215.00
0.2001 0.1163 0.1017
0.511996 0.511175 0.510791
1.67 1.98
Shiraishi et al. 1997b Black et al. 1987 Black et al. 1987 55
(Continued )
AGE CONSTRAINTS IN EAST ANTARCTICA
69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104
Sample
108 109 110 111 112 113 114 115 116 117 118 119 120
78285023 78285024 78285027 80285043B 80285043M 77283498 77283554 45121408 45121504 45121505 45121507 451216-0 45121506
Napier 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142
Complex SS97021307 SS97021208-1 SS97021303B SS96122803B-1 A90021603B A90021604G 21601G 21602A 21602AB 21602AW 21602B 21602C 21603C 21603E 21603H 21603I 21603N 21603G 76283267 77283464 77283465 77283466
Rock type
Locality
Nd Sm (ppm) (ppm)
147
Sm/144Nd
143
Nd/144Nd T(DM) Ga
References
Granitic orthogneiss Granite Gr orthogneiss Anorthosite layer Gabbroic layer Tonalitic orthogneiss Granitic orthogneiss Opx–grt gneiss Grt–biotite gneiss Opx–grt gneiss Fine grt –biotite gneiss Opx–grt gneiss Opx–grt gneiss
Mt. Fletta Condon Hills Thala Hills Amphitheatre Lakes Amphitheatre Lakes Ward Nunataks Mt. Underwood Sandercock Nunataks Sandercock Nunataks Sandercock Nunataks Sandercock Nunataks Sandercock Nunataks Sandercock Nunataks
6.26 6.61 3.82 1.73 1.43 2.92 3.64 7.66 10.20 8.86 7.52 10.20 15.11
34.70 48.20 20.90 10.70 8.00 12.30 21.10 33.50 54.60 47.40 38.40 54.30 63.52
0.1090 0.0829 0.1104 0.0977 0.1080 0.1435 0.1043 0.1382 0.1129 0.1130 0.1184 0.1135 0.1438
0.510682 0.510462 0.510710 0.510692 0.510716 0.511311 0.511326 0.512039 0.511943 0.511977 0.511991 0.511916 0.512080
2.28 2.08 2.27 2.04 2.26 2.02 1.27 2.03 1.66 1.61 1.68 1.71 2.10
Black et al. Black et al. Black et al. Black et al. Black et al. Black et al. Black et al. This study This study This study This study This study This study
Granitic gneiss Psammitic gneiss Mafic granulite Sapphirine –quartz gneiss Grt–felsic gniess Fe–rich grt–px gneiss Mafic gneiss Mafic gneiss Mafic gneiss Mafic gneiss Mafic gneiss Mafic gneiss Mafic gneiss Felsic gneiss Felsic gneiss Felsic gneiss Ultramafic gneiss Ultramafic gneiss Paragneiss Leucogneiss Grt–bg. gneiss Metapelite
Mt. Riiser-Larsen Mt. Riiser-Larsen Mt. Riiser-Larsen Mt. Riiser-Larsen Tonagh Island Tonagh Island Tonagh Island Tonagh Island Tonagh Island Tonagh Island Tonagh Island Tonagh Island Tonagh Island Tonagh Island Tonagh Island Tonagh Island Tonagh Island Tonagh Island Mount Sones Mount Sones Mount Sones Mount Sones
32.00 3.68 2.45 6.90 8.89 1.61 3.12 11.10 12.70 3.96 5.16 5.14 1.71 1.66 1.18 3.34 1.50 1.61 5.84 1.10 1.96 61.61
162.00 22.60 7.01 50.40 46.20 7.45 12.40 54.70 59.90 21.80 21.10 21.10 7.99 10.00 11.60 27.90 5.63 7.45 31.89 6.27 15.29 96.39
0.1194 0.0984 0.2113 0.0827 0.1163 0.1306 0.1521 0.1226 0.1281 0.1098 0.1478 0.1472 0.1293 0.1003 0.0615 0.0723 0.1610 0.1306 0.1106 0.1060 0.7740 0.3866
0.510961 0.510697 0.512756 0.510464 0.510914 0.510944 0.511498 0.510977 0.511015 0.510660 0.511805 0.511585 0.511109 0.510609 0.509972 0.510128 0.511576 0.510944 0.50996 0.50986 0.50893 0.51433
3.43 3.14
Suzuki et al. 2006 Suzuki et al. 2006 Suzuki et al. 2006 Suzuki et al. 2001 Owada et al. 2001 Owada et al. 2001 Owada et al. 1994 Owada et al. 1994 Owada et al. 1994 Owada et al. 1994 Owada et al. 1994 Owada et al. 1994 Owada et al. 1994 Owada et al. 1994 Owada et al. 1994 Owada et al. 1994 Owada et al. 1994 Owada et al. 1994 Black & McCulloch Black & McCulloch Black & McCulloch Black & McCulloch
3.03 3.39 3.99 3.53 3.71 3.56 2.90 3.43 3.58 3.32 3.10 3.18 3.97 3.44 3.43 3.77
1987 1987 1987 1987 1987 1987 1987
K. SHIRAISHI ET AL.
No.
56
Table 15. Continued
1987 1987 1987 1987
77283467 78285007-A 78285001-F 78285007-J 78285008-9 M N J1 J5 50 53 54 51 56 57
Mafic granulite Tonalitic orthogneiss Tonalitic orthogneiss Tonalitic orthogneiss Paragneiss Charnockite Charnockite Charnockite Charnockite Leuconorite Leuconorite Leuconorite Gabbro Gabbro Gabbro
Central Dronning Maud Land 158 J1704 Felsic gneiss 159 J1838 Felsic gneiss 160 J1671 Felsic gneiss 161 J1795 Felsic gneiss 162 J1736 Augen gneiss 163 J1797 Augen gneiss 164 J1698 Metagranodiorite 165 J1695 Metagranodiorite 166 SR39B/16 167 SR7W/23 Metagabbro 168 SR4/23 S-type granite 169 MS4/23 S-type granite 170 MS2/23 Metaquartzite 171 MS5/23 Calc-silicate rock 172 SR 17/14 Metanoritic dyke 173 SR 28D/16 Metanoritic dyke 174 13A-2/16 Enderbitic gneiss Enderbitic gneiss 175 13A-4/16 Enderbitic gneiss 176 13A-6/16 177 13A-8/16 Enderbitic gneiss 178 13A-9/16 Enderbitic gneiss 179 13A-10/16 Enderbitic gneiss 180 13A-0/16 Enderbitic gneiss
Mount Sones Mount Sones Mount Sones Mount Sones Mount Sones Fyfe Hills Fyfe Hills Fyfe Hills Fyfe Hills Fyfe Hills Fyfe Hills Fyfe Hills Fyfe Hills Fyfe Hills Fyfe Hills
6.66 3.49 4.36 3.97 5.81 1.81 0.82 1.70 0.53 1.73 1.34 1.15 10.05 4.99 5.49
26.87 24.42 29.24 30.64 20.42 15.72 6.11 10.37 4.65 13.37 7.49 5.50 46.92 20.61 29.84
0.1498 0.0863 0.0901 0.0783 0.1719 0.0695 0.0811 0.0990 0.0688 0.0782 0.1081 0.1263 0.1294 0.1463 0.1112
0.51057 0.50901 0.50912 0.50886 0.51084 0.50913 0.50936 0.50979 0.50917 0.50939 0.50997 0.51032 0.51034 0.51067 0.51007
Orvinfjella Wohlthatmassiv Orvinfjella Wohlthatmassiv Orvinfjella Wohlthatmassiv Orvinfjella Wohlthatmassiv Orvinfjella Wohlthatmassiv Orvinfjella Wohlthatmassiv Orvinfjella Wohlthatmassiv Orvinfjella Wohlthatmassiv Schirmacher Hills Schirmacher Hills Schirmacher Hills Schirmacher Hills Schirmacher Hills Schirmacher Hills Schirmacher Hills Schirmacher Hills Schirmacher Hills Schirmacher Hills Schirmacher Hills Schirmacher Hills Schirmacher Hills Schirmacher Hills Schirmacher Hills
10.51 46.8 9.67 31.3 15.68 77.5 8.355 34.32 8.991 41.03 10.14 58.42 4.759 17.04 8.355 34.32 0.698 1.96 12.3 40.2 8.93 38.2 12.8 62.4 5.58 27.2 15.8 105 11.50 178.00 6.16 39.40 13.4 46.7 5.81 28 11.5 44.7 10.1 58.3 6.77 27.3 7.92 31.3 9.97 36.34
0.1352 0.1860 0.1218 0.1568 0.132 0.1046 0.1681 0.1466 0.2157 0.1850 0.1412 0.1239 0.1237 0.0907 0.0594 0.1435 0.1735 0.1254 0.1555 0.1047 0.1499 0.153 0.1659
0.512145 0.512708 0.512294 0.512314 0.512165 0.512348 0.512468 0.512278 0.512884 0.512765 0.512391 0.512491 0.512202 0.512141 0.511884 0.512791 0.512538 0.51233 0.512487 0.51228 0.512426 0.512462 0.512461
3.33 3.36 3.31 3.27 3.25 3.33 3.42 3.52 3.67 3.28
1.72 1.22 1.62 1.61 0.95 1.71 1.33 0.92 1.40 1.10
1.21 1.42 1.05 1.44 1.42
Black & McCulloch 1987 Black & McCulloch 1987 Black & McCulloch 1987 Black & McCulloch 1987 Black & McCulloch 1987 McCulloch & Black 1984 McCulloch & Black 1984 McCulloch & Black 1984 McCulloch & Black 1984 McCulloch & Black 1984 McCulloch & Black 1984 McCulloch & Black 1984 McCulloch & Black 1984 McCulloch & Black 1984 McCulloch & Black 1984
Jacobs et al. 1998 Jacobs et al. 1998 Jacobs et al. 1998 Jacobs et al. 1998 Jacobs et al. 1998 Jacobs et al. 1998 Jacobs et al. 1998 Jacobs et al. 1998 Ravikant, 2007 Ravikant, 2007 Ravikant, 2007 Ravikant, 2007 Ravikant, 2007 Ravikant, 2007 Ravikant, 2006 Ravikant, 2006 Ravikant, 2004 Ravikant, 2004 Ravikant, 2004 Ravikant, 2004 Ravikant, 2004 Ravikant, 2004 Ravikant, 2004
147
Sm/144Nd , 0.13; remaining ages are 0.13
57
Rock types follow the original references. Depleted mantle Nd model ages (TDM), recalculated following Stern (2002). Bold numerals indicate TDM with 147 Sm/144Nd , 0.165. Mineral abbreviations are after Kretz (1983). Place names are shown in in Figures 2, 3 and 13.
4.18 3.95 3.94 3.88
AGE CONSTRAINTS IN EAST ANTARCTICA
143 144 145 146 147 148 149 150 151 152 153 154 155 156 157
58
K. SHIRAISHI ET AL.
Fig. 12. Histogram of depleted mantle model ages for the six terranes of eastern Dronning Maud Land and Enderby Land. Only samples yielding 147Sm/144Nd , 0.13 are plotted. Data are provided in Table 15.
southern Soˆya Coast region (Skallen, Skallevikshalsen and Rundva˚gshetta; see Fig. 3). Yoshida et al. (1999) and Suda et al. (2008) have suggested the presence of late Archaean to early Palaeoproterozoic crust in the southern part of the LHC. Such crust may represent an ancient basement to supracrustal lithologies, and/or source lithologies of detritus from the hinterland. TDM values of 12 samples from the SRM concentrate in a narrow range of 0.90–1.25 Ga, except one quartzofeldspathic gneiss (1.57 Ga) from the eastern part of the mountains. TDM values (1.0–1.1 Ma) for enderbitic gneiss in the
NE Terrane are only slightly older than the time of magmatism, estimated from SHRIMP zircon (951 + 17 Ma, this study), Sm –Nd whole-rock isochron (961 + 101 Ma) and Rb–Sr whole-rock isochron (978 + 52 Ma) ages (Shiraishi & Kagami 1992). TDM values of less than 1.0 Ga were obtained from younger granites in the SRM. The significance of 1.57 Ga gneiss from the eastern area (Balchenfjella) is not clear (Grew et al. 1992). Pan-African post-orogenic granites have only slightly younger TDM values(0.9–1.0 Ga) than orthogneisses (as recalculated from Arakawa et al. 1994). Although the number and variety of samples are limited, the
Fig. 13. Distribution of TDM in east Dronning Maud Land. Only samples with 147Sm/144Nd , 0.13 are plotted. Data sources are given in Table 15.
AGE CONSTRAINTS IN EAST ANTARCTICA
data indicate that the majority of the SRM is built on a basement of c. 1100–1000 Ma juvenile crust. This is in a good agreement with c. 1130 and c. 1000 Ma signatures from magmatic zircon in orthogneisses and detrital zircon in paragneisses, and supports the observation that there is no significant contribution from Palaeoproterozoic or Archaean continental crust. To the west of the SRM, TDM values reported by Jacobs et al. (1998), Ravikant et al. (2004, 2007) and Ravikant (2006) are consistently Mesoproterozoic. Recalculation in this study yields slightly younger results, with TDM values ranging from 1.4 to 0.9 Ga that are comparable with those obtained in the SRM. Thus the central Dronning Maud Land crust formed at similar times to that of the SRM.
Discussion and conclusions Tectonothermal events in the Sør Rondane Mountains Integrating the results of the SHRIMP dating and depleted mantle Nd model ages (TDM) with previous geochronological studies, tectonothermal events in the Sør Rondane Mountains and other terranes in central –eastern Dronning Maud Land are summarized in Table 16. Coincident ages of extensive magmatism and TDM values suggest that the basement crust of the SRM formed during the late Mesoproterozoic, as pointed out by Grew et al. (1992) and Shiraishi & Kagami (1992). Petrochemical studies of mafic to felsic orthogneiss indicate formation in oceanic and island-arc to continental margin environments (Osanai et al. 1992; Ikeda & Shiraishi 1998). There is little indication of high-grade metamorphism during c. 1130 and 1000 Ma magmatism, and, significantly, the data provide no clear evidence of a Grenvillian orogenic event produced by continent–continent collision. Mesoproterozoic juvenile crust has also been reported from the Lurio Belt of northern Mozambique (e.g. Grantham et al. 2008), the Vijayan Complex of Sri Lanka (e.g. Kro¨ner et al. 2003) and other Gondwana fragments. Some workers have discussed the unity of these terranes during Gondwana formation (e.g. Ravikant et al. 2007; Grantham et al. 2008). Although it is far beyond of the scope of this paper, it is essential to compare these terranes with precise age data. It is characteristic that c. 750–800 Ma magmatism (and possible metamorphism) is recorded in detrital zircon from paragneisses in the NE Terrane. These detrital ages demonstrate that sedimentation occurred after c. 750 Ma in the NE Terrane. Because some of
59
the sediments have volcaniclastic characteristics, it is also possible that sedimentation on a c. 1000 Ma basement occurred before and at 800–750 Ma. In any case, the formation of high-grade metamorphism occurred after 750 Ma. The abundant and widespread growth of metamorphic zircon at c. 600 Ma is interpreted as timing the peak of granulite-facies metamorphism and ductile deformation in the NE Terrane. Grew et al. (1989) and Asami et al. (2007) suggested two stages of metamorphism; around 760–800 8C and 7–8 kbar for peak conditions of granulite-facies metamorphism, and around 500–600 8C for subsequent amphibolite-facies metamorphism. We suggest that the later metamorphism took place at c. 550–570 Ma, as indicated by minor zircon growth in paragneiss (sample 85020401C) and the emplacement of relatively undeformed granite dykes (sample 90112302A) at c. 550 Ma, which may have derived from the mobilization of anatectic melts originally produced during the earlier granulite-grade metamorphism. It is worth noting that the c. 600 Ma episode of zircon growth is not present in all samples, even those from adjacent localities, and that the timing of granulite-grade metamorphism could be easily missed with a smaller set of samples. A similar phenomenon was also discussed for the southern Indian ultrahigh-temperature granulite-facies rocks (Santosh et al. 2008). Only one published Sm–Nd mineral isochron age of 624 + 18 Ma, from enderbitic gneiss in the NE Terrane, corresponds to the metamorphic zircon age (Shiraishi & Kagami 1992). Although the zircon results broadly support the assertion of Asami et al. (2005) that the granulitegrade metamorphic terrane was produced by a Pan-African event, it is difficult to reconcile the dominance of c. 600 Ma ages from metamorphic zircon with the dominance of c. 540– 515 Ma ages from metamorphic monazite. The discrepancy is present even in single samples of paragneiss: sample 85020401C, with metamorphic zircon ages of 609 + 11 Ma and 565 + 7 Ma, yielded a monazite age of 541 + 15 Ma (five unzoned grains); and sample 8402004, with no metamorphic zircon age identified, yielded monazite ages of 517 + 14 Ma (cores of 11 grains) and 542 + 12 Ma (rims of 11 grains; Asami et al. 2005). In another sample, two zoned monazite grains in garnet porphyroblasts yielded rim ages of 553 + 11 and 544 + 17 Ma and core ages of 753 + 37 and 697 + 47 Ma (Asami et al. 2005). Isotopic resetting of monazite through Pb loss at temperatures below granulite facies is now widely regarded as unlikely (Cherniak et al. 2004), so the younger monazite ages should be attributed to either new metamorphic growth or recrystallization. Because the emplacement of c. 550 Ma granitic dykes places a lower age
60
Table 16. SHRIMP and TDM ages from central–eastern Dronning Maud Land Central DML
Schirmacher Hills
Sør Rondane Mountains SW
c. 510 Ma low-temperature hydrothermal processes
Western Rayner Complex
c. 520 –550 Ma Peak metamorphism (UHT)
c. 520– 540 Ma Metamorpshim
515 Ma metamorphism (Ttn), post-orogenic granite c. 530 Ma Magmatism c. 560 Ma Metamorphism, Magmatism
c. 605 Ma late-kinematic grt –mus pegmatite c. 625 Ma c. 650 Ma Granulite-facies Magmatism metamorphism (UHT?)
c. 540 Ma Metamorphism c. 600– 650 Ma Peak granulite facies (UHT?)
c. 620 Ma Magmatism? c. 660 Ma Metamorphism?
c. 800 Ma Metamorphism/ magmatism?
c. 800 Ma Magmatism (det), metamorphism? (det) c. 1100 Magmatism and granulite-facies metamorphism
c. 1000 Ma Metamorphism
c. 1000 Ma Magmatism (prot þ det) c. 1130 Ma Magmatism
c.1000 Ma Magmatism(prot)
c. 1200, 3260 Ma (det) 0.9 –1.0, 1.6 Ga Crustal inheritance (TDM)
Rayner Complex
NE
c. 550 Ma Lamprophyre c. 575 –590 Ma Cooling
Lu¨tzow– Holm Complex
0.9– 1.0, 1.6 Ga Crustal inheritance (TDM)
K. SHIRAISHI ET AL.
c. 530 –515 Ma Granulite-facies metamorphism c. 570 Ma Amphibolite-facies metamorphism c. 600 Ma Magmatism
Yamato– Belgica Complex
c. 2470 (inh) 1.0 –1.8, 2.3 –2.7 Ga Crustal inheritance (TDM)
c. 2900 – 1500 Ma (det)
c. 910 –980 Ma Metamorphism, magmatism c. 1040 –2440 Ma (det) 2.3 Ga Crustal inheritance (TDM)
c. 1810 –2580 (det) 1.6 –2.3 Ga Crustal inheritance (TDM)
Bold numbers are SHRIMP zircon or titanite (Ttn) ages (inh, inherited; prot, protolith; det, detrital). TDM, depleted mantle model age (Ga); UHT, ultrahigh-temperature metamorphism. Ages from central Dronning Maud Land are after Jacobs et al. (1998) and Henjes-Kunst (2004). Those from Schirmacher Hills are based on Sm –Nd data after Ravikant (2006). Other data sources are Shiraishi et al. (1997b, 2003, and this study).
AGE CONSTRAINTS IN EAST ANTARCTICA
constraint on high-strain deformation and the formation of granulite-grade gneissic fabrics, the ages may be an indication that amphibolite-facies conditions (with monazite and garnet growth) persisted after 550 Ma. Monazite growth and/or recrystallization may also have occurred during contact metamorphism in the vicinity of late to post-tectonic intrusions of granitic and syenitic plutons. This magmatic activity can also be invoked to explain the c. 517 Ma age of titanite in sample 90102801A, through new titanite growth or closure of Pb diffusion at 700 –600 8C (Cherniak 1993). Further monazite dating in a detailed petrographic context is required to resolve these issues. The geochronological results from the SW Terrane are different, and shed some light on the polyphase nature of Pan-African metamorphism in the NE Terrane. The extensive meta-tonalites, produced by subduction-related magmatism (Ikeda & Shiraishi 1998), were emplaced at 960– 920 Ma (Takahashi et al. 1990, and this study). The relationship of meta-tonalites to adjacent lithologies is unclear, as they are bounded by the mylonitic Main Shear Zone, the timing of which is constrained at or later than c. 560 Ma by the intrusion of mylonitized granite at Vengen. Gneissic fabrics in the SW Terrane, associated with amphibolitefacies metamorphic assemblages, are temporally constrained between the ages of magmatic zircon (c. 650 Ma) and metamorphic zircon (c. 570 Ma) in a granitic orthogneiss. There is an absence of c. 600 Ma zircon in the SW Terrane, so that the c. 570 Ma age provides the best estimate of amphibolite-facies metamorphism and ductile deformation. This age coincides with our estimate of the timing of amphibolite-facies metamorphism in the NE Terrane, and unifies the metamorphic history of the terranes at this time. There is little evidence that the SW Terrane experienced high-grade metamorphism before 570 Ma, suggesting that the terranes have different origins. Scant c. 1130 Ma and c. 980 Ma age data from inherited zircon may tentatively indicate common crustal sources for both terranes, but this is far from proven. A unified geological history after 570 Ma is also supported by the presence of post-tectonic granitoids in both terranes, along with c. 517 Ma ages from metamorphic titanite and 500 –420 Ma K –Ar and Rb –Sr ages from various localities across the SRM (e.g. Shiraishi et al. 1997a). Lamprophyric and doleritic dykes, which have been metamorphosed to amphibolite-facies grade, intrude throughout the NE and SW Terranes in the western and central SRM, probably during the waning stages of c. 570 –560 Ma metamorphism. The geochemistry of the mafic dykes indicates a continental within-plate tectonic setting, with source magmas generated from a mixture of
61
subduction-related materials and metasomatically enriched mantle (Ikeda & Shiraishi 1995). A similar scenario was suggested for syenite magmatism in the Yamato Mountains (Zhao et al. 1995). Subsequent A-type granitic magmatism is the latest thermal event in the SRM (Li et al. 2003). Similar late to post-tectonic magmatism is also common in central to east Dronning Maud Land (e.g. Bauer et al. 2003; Roland 2004; Jacobs et al. 2008). In summary, supracrustal protoliths of the NE Terrane were deposited on a 1130–1000 Ma juvenile basement, at least partially after 750 Ma, and were metamorphosed under granulite-facies conditions during orogenesis at c. 600 Ma. Subsequent retrograde metamorphism took place at c. 560 Ma, synchronous with the juxtaposition of the NE and SW Terranes by large-scale shear zones that evolved into the Sør Rondane Suture and the mylonitic Main Shear Zone. The 560 Ma event may relate to the extensional collapse of the orogen, whereas the 600 Ma event relates to the peak collisional event. There is no evidence indicating whether the two events are in the same orogenic cycle or not. The 560 Ma event was followed by alkaline magmatism, which induced hightemperature contact metamorphism between 550 and 510 Ma. Cooling and sporadic magmatism continued over a protracted interval, possibly until as late as 420 Ma (Shiraishi et al. 1997a).
Comparing the SRM with neighbouring terranes Lu¨tzow-Holm, Rayner and Yamato – Belgica Complexes. TDM values from the Lu¨tzow-Holm Complex (LHC) suggest that para- and orthogneisses were derived from a variety of Mesoproterozoic and older basement lithologies. The southern Soˆya Coast differs from other parts of the LHC, with indications of Archaean and/or earlyPalaeoproterozoic basement. Areas with similarly ancient crustal residence ages are found in the Rayner and Napier Complexes in East Antarctica, as well as the Highland Complex in Sri Lanka. It is likely that these components relate to an unknown extent of ancient crust, hidden under the ice sheets of East Antarctica. The tectonic significance of Cape Hinode of LHC is controversial. TDM values (1.0–1.1 Ga) of meta-trondhjemite from Cape Hinode are almost contemporaneous with the crystallization age of magmatic zircon (c. 1017 Ma) and Sm –Nd wholerock isochron ages (c. 1030 Ma; Shiraishi et al. 1994, 1995). The meta-trondhjemite is characterized by relatively high Al and Sr, and low Y and Yb contents, similar to Archaean trondhjemites
62
K. SHIRAISHI ET AL.
and adakites. Trace element modelling suggests that parental magmas were derived by partial melting of a mid-ocean ridge basalt (MORB) source under garnet-stable P–T conditions (Ikeda et al. 1997; Hiroi et al. 2008). Thus both geochemical and geochronological evidence indicate a non-continental source for the c. 1000 Ma magmatism at Cape Hinode. In contrast, a paragneiss sample from Cape Hinode with TDM of 1.8 Ga, comparable with ages from Sandercock Nunataks in the Rayner Complex, indicates the involvement of older crustal sources in the sedimentary provenance. These various lines of evidence from the LHC are consistent with the gradual development of a continental margin throughout, and possibly after, the Mesoproterozoic. Hokada & Motoyoshi (2006) obtained electron microprobe CHIME ages and REE signatures for monazite from metapelitic granulites of Skallen in the LHC. They discovered a two-stage growth of monazite in two out of four samples, at 560– 500 Ma and 650 –580 Ma. They interpreted the older monazite growth as predating the peak metamorphism, on a prograde stage of the P –T path. How this two-stage metamorphism relates to orogenic events in the LHC has not been established, and requires further work. In the Rayner Complex, it has been known that gneisses represent granulite-grade reworking of the Archaean Napier Complex. However, inherited zircon ages and TDM values indicate that there is only minor contribution of Archaean crust to the Rayner Complex, in accordance with Black et al. (1987). The age of crustal protoliths to granulites in the Rayner Complex is still poorly understood. From a combination of SHRIMP U –Pb zircon and Sm– Nd mineral isochron ages, Shiraishi et al. (1997b) revealed a Pan-African aged overprint on rocks from the coastal regions of the western Rayner Complex. In contrast, inland outcrops of the Nye Mountains and Sandercock Nunataks yield zircon ages of 900– 1000 Ma. The Pan-African overprint of the Rayner Complex was also recognized by Motoyoshi et al. (2006), in CHIME monazite ages from metapelitic granulite of Forefinger Point. These monazites yield 750– 1000 Ma core ages with 517 –528 Ma rims. The rim ages correspond to SHRIMP U – Pb zircon ages of 530 –537 Ma (Shiraishi et al. 1997b). Thus the western Rayner Complex is geochronologically heterogeneous, with the coastal region adjacent to the LHC overprinted by a Cambrian orogeny. The Yamato –Belgica Complex (YBC) lies between the SRM and the LHC, and plays a critical role in understanding the relationship between the two regions. The Yamato Mountains are composed of large volumes of c. 500 Ma syenite (Zhao
et al. 1995). It has been suggested that syenite plutons formed within the hinterland of a c. 530–550 Ma continental collision zone, with the generation of syenite parental magma by partial melting in a mantle wedge above a subduction zone (Shiraishi et al. 1994; Zhao et al. 1995). Considering the TDM and lithological types present in the LHC, the subducting plate included Neoproterozoic continental margin sediments associated with a wide range of protolith and provenance ages. Shiraishi et al. (1994, 2003) showed for the first time that East Antarctica was not a united continent before the amalgamation of East and West Gondwana during the Pan-African tectonic event. Central and western Dronning Maud Land. In recent years, multiple stages of thermal events during the Neoproterozoic to Cambrian PanAfrican orogeny have been well documented in central and western Dronning Maud Land (e.g. Jacobs et al. 1998; Paulsson & Austrheim 2003; Bisnath et al. 2006). It has been suggested that the tectonothermal history of the SRM is similar to that of central Dronning Maud Land (DML) (e.g. Jacobs & Thomas 2002). Felsic magmatism took place at c. 1130 Ma in central to western Dronning Maud Land, (as described by Jacobs et al. (1998) and Bisnath et al. (2006), and has been related by these workers to the formation of an extensive volcanic arc. TDM values for meta-igneous rocks from Wohlthatmassiv and Orvinfjella in central DML vary from 1.02 to 1.74 Ga, assuming a two-stage evolution of the Sm–Nd isotope system. This age range is similar to results recalculated by the method of Stern (2002), with a range of 0.9–1.7 Ga that is in good agreement with TDM values for the SRM (0.9–1.6 Ga). Jacobs et al. (1998) concluded that the Grenville-age crust in DML is basically juvenile, and did not involve significant amount of previous crust. This also appears to be the case in the SRM. However, the present study reveals differences between the two regions. Mesoproterozoic (c. 1000 Ma) metamorphism in the SRM is not yet confirmed by zircon growth, although this may be an artefact of insufficient SHRIMP data. Two stages of late Neoproterozoic–Cambrian metamorphism are recognized in both central DML (Jacobs et al. 1998) and the SRM, but the timing is not identical: 570– 550 Ma and 530–515 Ma in Wohlthatmassiv and Orvinfjella, v. c. 650– 600 Ma and c. 570–550 Ma in the SRM. In contrast, Henjes-Kunst (2004) reported granulite-facies metamorphism at c. 625 Ma and cooling below c. 5008C after amphibolite-facies metamorphism at c. 575–590 Ma in the coastal Schirmacher Hills. This corresponds closely to the inferred timing of
AGE CONSTRAINTS IN EAST ANTARCTICA
similar events in the SRM. Ultrahigh-temperature metamorphism was reported from the Schirmacher Hills (Baba et al. 2006). Baba et al. (2008) have also reported contrasting P –T–t paths between the Schirmacher Hills and the inland mountains of central DML. They suggested that an isobaric cooling path in Schirmacher Hills is comparable with the granulite- to amphibolite-grade retrograde path for the SRM proposed by Asami & Shiraishi (1987) and Asami et al. (2007). Although c. 1000 Ma metamorphism has not been demonstrated in the SRM, the metamorphic conditions, P–T paths and ultrahigh-temperature metamorphism are comparable between the NE Terrane of the SRM and Schirmacher Hills. In contrast, the lack of c. 600 Ma granulite-grade metamorphism in the SW Terrane is comparable with inland mountains in central Dronning Maud Land, where c. 580 –550 Ma granulite-grade metamorphism occurs. However, further petrological studies tied to geochronology are required. In this context, future zircon chronology to constrain the age of sedimentation of supracrustal rocks in Schirmacher Hills is necessary to establish these regional relationships. In the present discussion we have focused on the crustal development and subsequent metamorphism of terranes in eastern Dronning Maud Land. The age signatures of each terrane in eastern DML are clearly distinguishable (Table 16). Basement rocks of the YBC may represent the easternmost part of the SRM terrane, and the tectonic setting of the SRM is comparable with that in central DML. In contrast to the SRM, protoliths of the LHC have a more complicated nature, and may represent a collage of terranes involving Archaean protoliths in the southern part of Lu¨tzow-Holm Bay and Proterozoic volcanic arc and oceanic components along the Prince Olav Coast. The late Neoproterozoic to Cambrian reworking of the western margin of the Mesoproterozoic Rayner Complex and the presence of Mesoproterozoic crustal signatures in the LHC suggests that the two terranes were associated prior to orogenesis during the Pan-African event. This event is polyphase, involving c. 600 Ma orogenesis in Dronning Maud Land and a final collision at c. 550 –530 Ma in the LHC. From the extensive database of geochronological data obtained in this region of East Antarctica, a picture of the complex formation of this critical section of Gondwana is beginning to emerge. We are grateful to R. Armstrong, K. Misawa, H. Kaiden and O. Tachikawa for helping with SHRIMP analysis, and S. Ohno for making thin sections. The depleted mantle model age by DePaolo’s model was calculated with the Excel spreadsheet given by courtesy of R. J. Stern. Discussions with Y. Hiroi, G. H. Grantham,
63
Y. Motoyoshi, S. Baba, Y. Osanai and M. Owada are highly appreciated. We thank M. Santosh, R. Fuck and J. Jacobs for their constructive reviews and for improving the manuscript. This research was financially supported by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science No. 13440151 to K.S.
References A JISHI , H., K AWANO , Y., K AWAKAMI , T. & I KEDA , T. 2004. Geochronological study of post-metamorphic granite from Kasumi Rock, Lu¨tzow-Holm Complex, East Antarctica. Polar Geoscience, 17, 35– 44 A RAKAWA , Y., T AKAHASHI , Y. & T AINOSHO , Y. 1994. Nd and Sr isotope characteristics of the plutonic rocks in the Sør Rondane Mountains, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 7, 49–59. A SAMI , M. & S HIRAISHI , K. 1987. Kyanite from the western part of the Sør Rondane Mountains, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 1, 150–168. A SAMI , M., G REW , E. S. & M AKIMOTO , H. 1990. A staurolite-bearing corundum–garnet gneiss from the eastern Sør Rondane Mountains, Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 4, 22–40. A SAMI , M., O SANAI , Y., S HIRAISHI , K. & M AKIMOTO , H. 1992. Metamorphic evolution of the Sør Rondane Mountains, East Antarctica. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Science. Terra, Tokyo, 7 –15. A SAMI , M., G REW , E. S. & M AKIMOTO , H. 1993. Contrasting mineral associations between two wollastonite-bearing calc-silicate gneisses, eastern Sør Rondane Mountains, Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences. 6, 57–71. A SAMI , M., S UZUKI , K. & A DACHI , M. 1996. Monazite ages by the chemical Th– U–total Pb isochron method for pelitic gneisses from the eastern Sør Rondane Mountains, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 9, 49–64. A SAMI , M., S UZUKI , K. & A DACHI , M. 1997. Th, U and Pb data and CHIME dating of monazites from metamorphic rocks of the Rayner, Lu¨tzow-Holm, Yamato– Belgica and Sør Rondane Complexes, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 10, 130–152. A SAMI , M., S UZUKI , K. & G REW , E. S. 2005. Monazite and zircon dating by the chemical Th–U– total Pb isochron method (CHIME) from Alasheyev Bight to the Sør Rondane Mountains, East Antarctica: a reconnaissance study of the Mozambique suture in eastern Queen Maud Land. Journal of Geology, 113, 59– 82. A SAMI , M., G REW , E. S. & M AKIMOTO , H. 2007. Relict sapphirine þ kyanite and spinel þ kyanite associations in pyrope garnet from the eastern Sør Rondane Mountains, East Antarctica. Lithos, 93, 107–125. B ABA , S., O WADA , M., G REW , E. S. & S HIRAISHI , K. 2006. Sapphirine granulite from Schirmacher Hills, Central Dronning Maud Land. In: F U¨ TTERER , D. K.,
64
K. SHIRAISHI ET AL.
D AMASKE , D., K LEINSCHMIDT , G., M ILLER , H. & T ESSONSOHN , F. (eds) Antarctic Contributions to Global Earth Science. Springer, Berlin, 37–44. B ABA , S., O WADA , M. & S HIRAISHI , K. 2008. Contrasting metamorphic P –T path between Schirmacher Hills and Mu¨hlig-Hofmannfjella, central Dronning Maud Land, East Antarctica. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: a Key to the East–West Gondwana Connection. Geological Society, London, Special Publications, 308, 401 –418. B AUER , W., T HOMAS , R. J. & J ACOBS , J. 2003. Proterozoic—Cambrian history of Dronning Maud Land in the context of Gondwana assembly. In: Y OSHIDA , M., W INDLEY , B. F. & D ASGUPTA , S. (eds) Proterozoic East Gondwana: Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 247– 269. B ISNATH , A., F RIMMEL , H. E., A RMSTRONG , R. A. & B OARD , W. S. 2006. Tectono-thermal evolution of the Maud Belt: New SHRIMP U– Pb zircon data from Gjelsvikfjella, Dronning Maud Land, East Antarctica. Precambrian Research, 150, 95– 121. B LACK , L. P. & M C C ULLOCH , M. T. 1987. Evidence for isotopic equilibration of Sm– Nd whole-rock systems in early Archean crust of Enderby Land, Antarctica, Earth and Planetary Science Letters, 82, 15– 24. B LACK , L. P., S HERATON , J. W. & J AMES , P. R. 1986. Late Archaean granites of the Napier Complex, Enderby Land, Antarctica: A comparison of Rb– Sr, Sm–Nd and U–Pb isotopic systematic in a complex terrain. Precambrian Research, 32, 343 –368. B LACK , L. P., H ARLEY , S. L., S UN , S. S. & M C C ULLOCH , M. T. 1987. The Rayner Complex of East Antarctica: complex isotopic systematics within a Proterozoic mobile belt. Journal of Metamorphic Geology, 5, 1 –26. B OGER , S. D., W ILSON , C. J. L. & F ANNING , C. M. 2001. Early Paleozoic tectonism within the East Antarctic craton: The final suture between east and west Gondwana? Geology, 29, 463 –466. B USHNELL , V. C. & C RADDOCK , V. C. (eds) 1969, 1970. Geologic maps of Antarctica. American Geographical Society, Antarctic Map Folio Series, Folio 12. C HERNIAK , D. J. 1993. Lead diffusion in titanite and preliminary results on the effects of radiation damage on Pb transport. Chemical Geology, 110, 177–194. C HERNIAK , D. J., W ATSON , E. B., G ROVE , M. & H ARRISON , T. M. 2004. Pb diffusion in monazite: a combined RBS/SIMS study. Geochimica et Cosmochimica Acta, 68, 829–840 C OLLINS , A. S., S ANTOSH , M., B RAUN , I. & C LARK , C. 2007. Age and sedimentary provenance of the Southern Granulites, South India: U –Th–Pb SHRIMP secondary ion mass spectrometry. Precambrian Research, 155, 125– 138. D E P AOLO , D. J. 1988. Neodymium Isotope Geochemistry. Springer, New York. D ICKIN , A. P. 1995. Radiogenic Isotope Geology. Cambridge University Press, Cambridge. F ITZSIMONS , I. C. W. 2000. Grenville-age basement provinces in East Antarctica; evidence for three separate collisional orogens. Geology, 28, 879–882.
F RIMMEL , H. E. 2004. Formation of a late Mesoproterozoic supercontinent: the South Africa–East Antarctica connection. In: E RIKSSON , P. G., A LTERMANN , W., N ELSON , D. R., M UELLER , W. O. & C ATUNEANU , O. (eds) The Precambrian Earth: Tempos and Events. Developments of Precambrian Geology, 12, 240– 255. G EISLER , T., P IDGEON , R. T., K URTZ , R., V AN B RONSWIJK , W. & S CHLEICHER , H. 2003. Experimental hydrothermal alteration of partially metamict zircon. American Mineralogist, 88, 1496– 1513. G RANTHAM , G. H., M ABOKO , M. & E GLINGTON , B. M. 2003. A review of the evolution of the Mozambique Belt and implications for the amalgamation of Rodinia and Gondwana. In: Y OSHIDA , M., W INDLEY , B. F. & D ASGUPTA , S. (eds) Proterozoic East Gondwana, Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 401–426. G RANTHAM , G. H., M ACEY , P. H., I NGRAM , B. A. ET AL . 2008. Terrane correlation between Antarctica, Mozambique and Sri Lanka; comparisons of geochronology, lithology, structure and metamorphism and possible implications for the geology of southern Africa and Antarctica. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctic: a Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 91–120. G REW , E. S. 1982. The Antarctic margin. In: A LAN , E. M. N. & S TEHLI , F. G. (eds) The Ocean Basins and Margins, Plenum Publishing Corporation, New York, 6, 697–755. G REW , E. S., A SAMI , M. & M AKIMOTO , H. 1989. Preliminary petrological studies of the metamorphic rocks of the eastern Sør Rondane Mountains, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 3, 100– 127. G REW , E. S., M ANTON , W. I., A SAMI , M. & M AKIMOTO , H. 1992. Geochronologic data on Proterozoic polymetamorphic rocks of the eastern Sør Rondane Mountains, East Antarctica. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Science. Terra, Tokyo, 37–44. H ENJES -K UNST , F. 2004. Further evidence for PanAfrican polyphase magmatism and metamorphism in central Dronning Maud Land, East Antarctica, from rocks at Schirmacheroase: A geochronological study. Geologisches Jahrbuch, Reihe B, 96, 255– 291. H IROI , Y., S HIRAISHI , K. & M OTOYOSHI , Y. 1991. Late Proterozoic paired metamorphic complexes in East Antarctica, with special reference to the tectonic significance of ultramafic rocks. In: T HOMSON , M. R.A., C RAME , J. A. & T HOMSON , J. W. (eds) Geological Evolution of Antarctica. Cambridge University Press, Cambridge, 83–87. H IROI , Y., M OTOYOSHI , Y., I SHIKAWA , N., H OKADA , T. & S HIRAISHI , K. 2008. Origin of xenocrystic garnet and kyanite in clinopyroxene –hornblende-bearing adakitic meta-tonalites from Cape Hinode, Prince Olav Coast, East Antarctica. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: a Key to the East– West Gondwana
AGE CONSTRAINTS IN EAST ANTARCTICA Connection. Geological Society, London, Special Publications, 308, 333– 352. H OKADA , T. & M OTOYOSHI , Y. 2006. Electron microprobe technique for U– Th–Pb and REE chemistry of monazite, and its implications for pre-, peak- and post-metamorphic events of the Lu¨tzow-Holm Complex and the Napier Complex, East Antarctica. Polar Geoscience, 19, 118– 151. I KEDA , Y. & S HIRAISHI , K. 1995. Geochemical characteristics of mafic dike rocks from the Sør Rondane Mountains, East Antarctica. Polar Geoscience, 8, 49– 64. I KEDA , Y. & S HIRAISHI , K. 1998. Petrogenesis of the tonalitic rocks from the Sør Rondane Mountains, East Antarctica. Polar Geoscience, 11, 143– 153. I KEDA , Y., S HIRAISHI , K. & Y ANAI , K. 1997. Petrogenesis of the meta-trondhjemites from Cape Hinode, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 10, 102– 110. I SHIZUKA , H., S UZUKI , S. & K OJIMA , H. 1995. Mineral paragenesis of the sapphirine-bearing rock from Austkampane area of the Sør Rondane Mountains, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 8, 65–74 I SHIZUKA , H., S UZUKI , S. & K OJIMA , H. 1996. Meta-ultramafic rock from the Austkampane area of the Sør Rondane Mountains, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 9, 40– 48. J ACOBS , J. & T HOMAS , R. J. 2002. The Mozambique Belt from an East Antarctic perspective. In: G AMBLE , J. A., S KINNER , D. N. B. & H ENRYS , S. (eds) Antarctica at the Close of a Millennium. Royal Society of New Zealand Bulletin, 35, 3– 18. J ACOBS , J. & T HOMAS , R. J. 2004. A Himalayan-type indenter-escape tectonic model for the southern part of the Late Neoproterozoic/Early Paleozoic East African– Antarctic Orogen. Geology, 32, 721– 724. J ACOBS , J., F ANNING , C. M., H ENJES -K UNST , F., O LESH , M. & P AECH , H.-J. 1998. Continuation of the Mozambique Belt into East Antarctica: Grenville age metamorphism and polyphase Pan-African high grade events in Central Dronning Maud Land. Journal of Geology, 106, 385–406. J ACOBS , J., F ANNING , C. M. & B AUER , W. 2003a. Timing of Grenville-age vs. Pan-African medium- to high-grade metamorphism in western Dronning Maud Land (East Antarctica) and significance for correlations in Rodinia and Gondwana. Precambrian Research, 125, 1 –20. J ACOBS , J., K LEMD , J. R., F ANNING , C. M., B AUER , W. & C OLOMBO , F. 2003b. Extensional collapse of the Late Neoproterozic– Early Palaeozoic East African– Antarctic Orogen in central Dronning Maud Land, East Antarctica. In: Y OSHIDA , M., W INDLEY , B. F. & D ASGUPTA , S. (eds) Proterozoic East Gondwana: Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 271–287. J ACOBS , J., B INGEN , B., T HOMAS , R. J., B AUER , W., W INGATE , M. T. D. & F EITIO , P. 2008. Early Paleozoic orogenic collapse and voluminous latetectonic magmatism in Dronning Maud Land and Mozambique: insights into the partially delaminated
65
orogenic root of the East African–Antarctic Orogen? In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: a Key to the East–West Gondwana Connection. Geological Society, London, Special Publications, 308, 69–90. K AGAMI , H., I WATA , M., S ANO , S. & H ONMA , H. 1987. Sr and Nd isotopic compositions and Rb, Sr, Sm and Nd concentrations of standard samples. Technical Reports, Institute of Study of Earth’s Interior, Okayama University, Series 4, 1 –16. K AGAMI , H., K AWANO , Y., A KIYAMA , M. ET AL . 2006. Provenance of Paleozoic –Mesozoic sedimentary rocks in the Inner Zone of Southwest Japan: An evaluation based on Nd model ages. Gondwana Research, 9, 142– 151. K AWANO , Y., M ENO , A., N ISHI , N. & K AGAMI , H. 2005. Geochemistry of the pre/syn-metamorphic granite in the Ongul Islands, East Antarctica. Polar Geoscience, 18, 114– 129. K INNY , P. D., M C N AUGHTON , N. J., F ANNING , C. M. & M ASS , R. 1994. 518 Ma sphene (titanite) from the Khan Pegmatite, Namibia, Southwest Africa: a potential ion microprobe standard. US Geological Survey, Circular, 1107, 171. K RETZ , R. 1983. Symbols for rock-forming minerals. American Mineralogist, 68, 277– 279. K RO¨ NER , A., K EHELPANNALA , K. V. W. & H EGNER , E. 2003. Ca. 750– 1100 Ma magmatic events and Grenville-age deformation in Sri Lanka: relevance for Rodinia supercontinent formation and dispersal, and Gondwana amalgamation. Journal of Asian Earth Sciences, 22, 279 –300. L I , Z., T AINOSHO , Y., K IMURA , J., S HIRAISHI , K. & O WADA , M. 2003. Pan-African alkali granitoid from the Sør Rondane Mountains, East Antarctica. Gondwana Research, 6, 595– 605. L UDWIG , K. R. 2001. SQUID 1.00 User’s Manual. Berkeley Geochronology Center, Special Publications, 2. L UDWIG , K. R. 2003. Isoplot/Ex 3.0 User’s Manual. Berkeley Geochronology Center, Special Publications, 4. M C C ULLOCH , M. T. & B LACK , L. P. 1984. Sm– Nd isotopic systematics of Enderby Land granulites and evidence for the redistribution of Sm and Nd during metamorphism. Earth and Planetary Science Letters, 71, 46–58. M EERT , J. G. 2001. Growing Gondwana and rethinking Rodinia: a paleomagnetic perspective. Gondwana Research, 4, 279–288. M EERT , J. G. 2003. A synopsis of events related to the assembly of eastern Gondwana. Tectonophysics, 362, 1– 40. M EERT , J. G. & L IEBERMAN , B. S. 2007. The Neoproterozoic assembly of Gondwana and its relationship to the Ediacaran–Cambrian radiation. Gondwana Research, 14, 5–21. M IKHALSKY , E. V., B ELIATSKY , B. V., S HERATON , J. W. & R OLAND , N. W. 2006. Two distinct Precambrian terranes in the Southern Prince Charles Mountains, East Antarctica: SHRIMP dating and geochemical constraints. Gondwana Research, 9, 291 –309. M OTOYOSHI , Y. & I SHIKAWA , M. 1997. Metamorphic and structural evolution on granulites from Rundva˚gshetta, Lu¨tzow-Holm Bay, East Antarctica. In:
66
K. SHIRAISHI ET AL.
R ICCI , C. L. (ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antartica, Sienna, 65–72. M OTOYOSHI , Y., H OKADA , T. & S HIRAISHI , K. 2006. Electron microprobe (EMP) dating on monazite from Forefinger Point granulites, East Antarctica. In: F U¨ TTERER , D. K., D AMASKE , D., K LEINSCHMIDT , G., M ILLER , H. & T ESSENSOHN , F. (eds) Contributions to Global Earth Sciences. Springer, Berlin, 63–68. N ELSON , B. K. & D E P AOLO , D. J. 1984. Rapid production of continental crust 1.7 to 1.9 b.y. ago: Nd isotopic evidence from the basement of the North American mid-continent. Geological Society of America Bulletin, 96, 746– 754. N ISHI , N., K AWANO , Y. & K AGAMI , H. 2002. Rb –Sr and Sm–Nd isotopic geochronology of the granitoid and hornblende biotite gneiss from Oku-iwa Rock in the Lu¨tzow-Holm Complex, East Antarctica. Polar Geosciences, 15, 46–65. O SANAI , Y., S HIRAISHI , K., T AKAHASHI , Y. ET AL . 1992. Geochemical characteristics of the metamorphic rocks from the Central Sør Rondane Mountains, East Antarctica. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Science. Terra, Tokyo, 17–27. O WADA , M., O SANAI , Y. & K AGAMI , H. 1994. Isotopic equilibration age of Sm– Nd whole-rock system in the Napier Complex (Tonagh Island), East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 7, 122– 132. O WADA , M., O SANAI , Y., T SUNOGAE , T., H AMAMOTO , T., K AGAMI , H., T OYOSHIMA , T. & H OKADA , T. 2001. Sm– Nd garnet ages of retrograde garnet bearing granulites from Tonagh Island in the Napier Complex, East Antarctica: A preliminary study. Polar Geosciences, 14, 75–87. P ACES , J. B. & M ILLER , J. D., J R 1993. Precise U–Pb ages of Duluth Complex and related mafic intrusions, northeastern Minnesota: Geochronological insights to physical, petrogenetic, paleomagnetic, and tectonomagmatic processes associated with the 1.1 Ga Midcontinent Rift System. Journal of Geophysical Research, 98, 13997– 14013. P AULSSON , O. & A USTRHEIM , H. 2003. A geochronological and geochemical study of rocks from Gjelsvikfjella, Dronning Maud Land, Antarctica— implications for Mesoproterozoic correlations and assembly of Gondwana. Precambrian Research, 125, 113– 138. P ICCIOTTO , E., D EUTCH , S. & P ASTEELS , P. 1964. Isotopic ages from the Sør Rondane Mountains, Dronning Maud Land. In: A DIE , R. J. (ed.) Antarctic Geology. North-Holland, Amsterdam, 570–578. R AVICH , M. G. & K AMENEV , Ye. N. 1975. Crystalline Basement of the Antarctic Platform. Gidrometeoizdat, Leningrad. R AVIKANT , V. 2006. Sm– Nd isotopic evidence for Late Mesoproterozoic metamorphic relics in the East African Orogen from the Schirmacher Oasis, East Antarctica. Journal of Geology, 114, 615– 625. R AVIKANT , V., B HASKAR R AO , Y. J. & G OPALAN , K. 2004. Schirmacher Oasis as extension of the Neoproterozoic East African Orogen into Antarctica:
new Sm– Nd isochron age constraints. Journal of Geology, 112, 607 –616. R AVIKANT , V., L AUX , J. H. & P IMENTEL , M. M. 2007. Sm– Nd and U– Pb isotopic constraints for crustal evolution during Late Neoproterozic from rocks of the Schirmacher Oasis, East Antarctica: geodynamic development coeval with the East African Orogeny. In: C OOPER , A. K., R AYMOND , C. R. & ISAES, Editorial Team (eds) Antarctica; A Keystone in a Changing World—Online Proceedings for the Tenth International Symposium on Antarctic Earth Sciences. US Geological Survey Open-File Report, 2007-1047 (DVD-ROM). World Wide Web Address: http:// pubs.usgs.gov/of /2007/1047/. R OLAND , N. W. 2004. Pan –African granite– charnockite magmatism in central Dronning Maud Land, East Antarctica: petrography, geochemistry and plate tectonic implications. Geologisches Jahrbuch Reihe B, 96, 187–231. S ANTOSH , M., C OLLINS , A. S., T AMASHIRO , I., K OSHIMOTO , S., T SUTSUMI , Y. & Y OKOYAMA , K. 2006. The timing of ultrahigh-temperature metamorphism in Southern India: U–Th– Pb electron microprobe ages from zircon and monazite in sapphirine-bearing granulites. Gondwana Research, 10, 128–155. S ANTOSH , M., T SUNOGAE , T., T SUTSUMI , Y. & I WAMURA , M. 2008. Microstructurally controlled monazite chronology of ultrahigh-temperature granulites from southern India: implications for the timing of Gondwana assembly. Island Arc (in press), doi:10.1111/j-1440-1738.2007.00601.X. S HIRAISHI , K. & K AGAMI , H. 1992. Sm–Nd and Rb–Sr ages of the metamorphic rocks from the Sør Rondane Mountains, East Antarctica. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Science. Terra, Tokyo, 29–35. S HIRAISHI , K. & K OJIMA , S. 1987. Basic and intermediate gneisses from the western part of the Sør Rondane Mountains, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 1, 129–149. S HIRAISHI , K., A SAMI , M., I SHIZUKA , H. ET AL . 1991. Geology and metamorphism of the Sør Rondane Mountains, East Antarctica. In: T HOMSON , M. R. A., C RAME , J. A. & T HOMSON , J. W. (eds) Geological Evolution of Antarctica. Cambridge University Press, Cambridge, 77 – 82. S HIRAISHI , K., H IROI , Y., E LLIS , D. J., F ANNING , C. M., M OTOYOSHI , Y. & N AKAI , Y. 1992. The first report of a Cambrian orogenic belt in East Antarctica—an ion microprobe study of the Lu¨tzow-Holm Complex. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Science. Terra, Tokyo, 67–73. S HIRAISHI , K., E LLIS , D. J., H IROI , Y., F ANNING , C. M., M OTOYOSHI , Y. & N AKAI , Y. 1994. Cambrian orogenic belt in East Antarctica and Sri Lanka: implications for Gondwana assembly. Journal of Geology, 102, 47–65. S HIRAISHI , K., K AGAMI , H. & Y ANAI , K. 1995. Sm– Nd and Rb –Sr isochron ages for meta-trondhjemites from Cape Hinode, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 8, 130–136.
AGE CONSTRAINTS IN EAST ANTARCTICA S HIRAISHI , K., O SANAI , Y., I SHIZUKA , H. & A SAMI , M. 1997a. Geological map of the Sør Rondane Mountains, Antarctica. Antarctic Geological Map Series, Sheet 35, scale 1 : 250 000. National Institute of Polar Research, Tokyo. S HIRAISHI , K., E LLIS , D. J., F ANNING , C. M., H IROI , Y., K AGAMI , H. & M OTOYOSHI , Y. 1997b. Reexamination of the metamorphic and protolith ages of the Rayner complex, Antarctica: Evidence for the Cambrian (Pan-African) regional metamorphic event. In: R ICCI , C. A. (ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antartica, Sienna, 79– 88. S HIRAISHI , K., H OKADA , T., F ANNING , C. M., M ISAWA , K. & M OTOYOSHI , Y. 2003. Timing of thermal events in the Dronning Maud Land, East Antarctica. Polar Geoscience, 16, 76–99. S TACEY , J. C. & K RAMERS , J. 1975. Approximation of terrestrial lead isotope evolution by a two-stage model. Earth and Planetary Science Letters, 26, 207–221. S TERN , R. J. 1994. Arc assembly and continental collision in the Neoproterozoic East Africa Orogen: Implications for the consolidation of Gondwana. Annual Review of Earth and Planetary Sciences, 22, 319–351. S TERN , R. J. 2002. Crustal evolution in the East African Orogen: a neodymium isotopic perspective. Journal of African Earth Sciences, 34, 109–117. SUDA, Y., K AWANO , Y., Y AXLEY , G., K ORENAGA , H. & H IROI , Y. 2008. Magmatic evolution and tectonic setting of metabasites from Lu¨tzow-Holm Complex, East Antarctica. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctic: a Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 211– 234.
67
S UZUKI , S., K AGAMI , H., I SHIZUKA , H. & H OKADA , T. 2001. Sm– Nd mineral isochron age of sapphirine – quartz gneiss from the Mt. Riiser-Larsen area in the Napier Complex, East Antarctica. Polar Geoscience, 14, 88–98. S UZUKI , S., A RIMA , M., W ILLIAMS , I. S., S HIRAISHI , K. & K AGAMI , H. 2006. Thermal history of UHT metamorphism in the Napier Complex, East Antarctica insights from zircon, monazite, and garnet ages. Journal of Geology, 114, 65– 84. T AKAHASHI , Y., A RAKAWA , Y., S AKIYAMA , T., O SANAI , Y. & M AKIMOTO , H. 1990. Rb– Sr and K– Ar whole rock ages of the plutonic bodies from the Sør Rondane Mountains, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 4, 1– 8. T INGEY , R. (ed.) 1991. The Geology of Antarctica. Oxford University Press, Oxford. Y OSHIDA , M., K AGAMI , H. & U NIIKRISHNAN W ARRIER , C. 1999. Neodymium model ages from Eastern Ghats and Lu¨tzow-Holm Bay: constraints on isotopic provinces in India– Antarctic sector of East Gondwana. In: R AO , A. T., D IVAKARA R AO , V. & Y OSHIDA , M. (eds) Eastern Ghats Granulites. Gondwana Research Group Memoir, 5, 161– 172. Y OSHIDA , M., J ACOBS , J., S ANTOSH , M. & R AJESH , H. M. 2003. Role of Pan-African events in the Circum-East Antarctic Orogen of East Gondwana: a critical overview. In: Y OSHIDA , M., W INDLEY , B. F. & D ASGUPTA , S. (eds) Proterozoic East Gondwana: Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 57–75. Z HAO , J.-X., S HIRAISHI , K., E LLIS , D. J. & S HERATON , J. W. 1995. Geochemical and isotopic studies of syenites from the Yamato Mountains, East Antarctica: Implications for the origin of syenitic magmas. Geochimica et Cosmochimica Acta, 59, 1363– 1382.
Early Palaeozoic orogenic collapse and voluminous late-tectonic magmatism in Dronning Maud Land and Mozambique: insights into the partially delaminated orogenic root of the East African – Antarctic Orogen? JOACHIM JACOBS1, BERNARD BINGEN2, ROBERT J. THOMAS3, WILFRIED BAUER3, MICHAEL T. D. WINGATE4 & PAULINO FEITIO5 1
Department of Earth Science, University of Bergen, Alle´gaten 41, 5007 Bergen, Norway (e-mail:
[email protected]) 2
Geological Survey of Norway, 7491 Trondheim, Norway
3
British Geological Survey, Kingsley Dunham Centre, Keyworth, Nottingham NG12 5GG, UK
4
Geological Survey of Western Australia, 100 Plain Street, East Perth, W.A. 6004, Australia 5
Direca˜o Nacional Geologia, Maputo, Mozambique
Abstract: The late tectonic history of the southern part of the Late Neoproterozoic– Early Palaeozoic East African–Antarctic Orogen (EAAO) is characterized by lateral extrusion, extensional collapse and large volumes of high-temperature A2-type granitoids. This late-tectonic igneous province covers an area more than 15 000 km2 of the EAAO in Dronning Maud Land (East Antarctica) and its northerly continuation as the Nampula Complex of NE Mozambique. The magmatic province is bounded in the north by the Lurio Belt. New secondary ionization mass spectrometry (SIMS) U–Pb analyses of zircons from two major late-tectonic granitoid intrusions from Dronning Maud Land indicate crystallization ages of 501 + 7 and 499 + 4 Ma, whereas a major extensional shear zone was dated at 507 + 9 Ma. New SIMS zircon U–Pb analyses of late-tectonic granitoid sheets and plutons from the Nampula Province indicate ages of 512 + 4, 508 + 4, 508 + 2 and 507 + 3 Ma. Consequently, the late-tectonic magmatism can be bracketed between c. 530 and 485 Ma. It started with small gabbro bodies emplaced at c. 530– 520 Ma, culminated with the intrusion of major granite– charnockite plutons at c. 510–500 Ma and terminated with the introduction of small volumes of sheet-like granite at c. 485 Ma. The new dates demonstrate that extensional shearing and granitoid intrusion are synchronous, and that orogenic collapse and the magmatism are related. We ascribe the distribution, structural style, geochemical composition and age of the late magmatic province to a process of partial delamination of the orogenic root in the southern third of the EAAO. It remains to be tested whether there is a relationship between orogenic collapse–granitoid magmatism and south-directed escape tectonics in the southernmost EAAO.
Northern Mozambique and Dronning Maud Land (East Antarctica) are interpreted to together represent the southern end of the Late Neoproterozoic– Early Palaeozoic East African–Antarctic Orogen (EAAO) (e.g. Stern 1994; Jacobs & Thomas 2004). This orogen stretches for more than 8000 km from Egypt–Arabia in the north, southwards through East Africa (including Madagascar) into northern Mozambique and thence into Dronning Maud Land in East Antarctica (Fig. 1). The EAAO resulted from a multi-plate collision of various parts of East and West Gondwana and shows a strong lateral variation in orogen style, probably as a result of this complex collision. It is characterized by accretion in its northern third (the ‘Arabian–Nubian Shield’) and
by continent–continent collision in its central and southern part. The deep erosion level exposed allows unique insights into an orogen that fundamentally changes in character along strike. The southernmost segment of the orogen, and its eventual termination, is recognized within Dronning Maud Land. This area is characterized by orogenic collapse and lateral extrusion tectonics, similar to the present situation in southeastern Asia, resulting from the collision of India and Asia (Jacobs & Thomas 2004). The southern third of the orogen from Antarctica into northern Mozambique is characterized by the intrusion of large volumes of A2-type granite (Roland 2004a, b), the volume of which decreases dramatically at the Lurio Belt, a conspicuous shear belt in
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 69– 90. DOI: 10.1144/SP308.3 0305-8719/08/$15.00 # The Geological Society of London 2008.
70 J. JACOBS ET AL. Fig. 1. (a) Geological setting of the East African– Antarctic Orogen (EAAO) after Jacobs & Thomas (2004). (b) The southern part of the EAAO affected by lateral extrusion and widespread intrusion of late-tectonic granitoids. The latter are largely confined to the EAAO and effectively terminate along the Lurio Belt in northern Mozambique. ANS, Arabian– Nubian Shield; C, Coats Land; DML, Dronning Maud Land; EH, Ellsworth– Haag; F, Filchner block; FI, Falkland Islands; G, Grunehogna; Ga, Gariep Belt; H, Heimefrontfjella; K, Kirwanveggen; L, Lurio Belt; Na---Na, Namaqua –Natal; SR, Shackleton Range; Da, Damara Belt; LH, Lu¨tzow-Holm Bay; M, Madagascar; Sa, Saldania Belt; Z, Zambezi Belt.
GRANITES IN EAAO
NE Mozambique, that trends oblique to the overall north–south strike of the EAAO. In this paper we review and highlight the significance of extensional shearing within the EAAO of northern Mozambique and Dronning Maud Land, and provide new age constraints for the extensional shear zones and associated late-tectonic granitoid intrusions in both areas.
Geological setting In Dronning Maud Land the Late Neoproterozoic – Early Palaeozoic collision along the EAAO to a large extent overprints Mesoproterozoic basement (Figs 1 and 2). It has been shown that this older crust is predominantly juvenile and was generated in island arcs along the margin of the ProtoKalahari Craton at c. 1.1 Ga (Jacobs et al. 1998, 2008; Bauer et al. 2003). The rocks underwent a first high-grade metamorphic event, associated with abundant syntectonic granitoids, between c. 1090 and 1070 Ma (Jacobs et al. 1998). This regional metamorphism was related to a collision event, which led to this area being incorporated into the supercontinent of Rodinia. After the
71
Mesoproterozoic orogenesis, there is little evidence for tectonic activity between c. 1050 and 650 Ma, with the exception of the Schirmacher Oasis area, where there is limited evidence for granitoid intrusion at c. 760 Ma (Jacobs et al. unpubl. data). The Late Neoproterozoic –Early Palaeozoic collision history can be separated into three major phases, as follows. (1) An earliest granulite facies stage is recorded in the Schirmacher Nappe at c. 620 Ma (HenjesKunst 2004), followed by anorthosite magmatism in the main mountain range at c. 600 Ma (Jacobs et al. 1998). This event is associated with shallowly inclined structures that are probably related to nappe emplacement, the age of which is unknown, but that might be related to the granulite- facies metamorphism. (2) The main deformation and medium- to highgrade metamorphism in the main mountain range is bracketed in age by metamorphic zircon rims between c. 590 and 550 Ma, and is interpreted to represent the collision phase (Jacobs et al. 1998, 2003b). It produced tight to isoclinal, upright east –west- to ESE–WNW-trending folds, which are post-dated by a major sinistral shear zone along the southern margin of the mountain range
Fig. 2. Overview map of western and central Dronning Maud Land, Antarctica, depicting main structural trends of the EAAO, extent of late-tectonic intrusions and sample localities. Western orogenic front of the EAAO is a major shear zone. A, Annandagstoppane; AH, Alexander von Humboldt Gebirge; B, Borgmassivet; F, Filchnerfjella; G, Gjelsvikfjella; H, Heimefrontfjella; MG, Mu¨hlig-Hofmann Gebirge; N, Novolazarevskaya Station; O, Orvinfjella; OG, Otto von Gruber Gebirge; S, Schirmacher Nappe; W, Wohlthatmassiv.
72
J. JACOBS ET AL.
in Orvinfjella and transpressive structures in the Wohlthatmassiv (Bauer et al. 2004). These structures are cut by discrete extensional shear zones and are intruded by undeformed pegmatites and granite veins, constraining a Cambrian age for this tectonic event (Jacobs et al. 2003b). (3) A late-tectonic stage is associated with extension, tectonic exhumation and south-directed crustal extrusion between c. 530 and 485 Ma, exposing mid- to lower crustal levels (e.g. Jacobs et al. 2003a; Engvik & Elvevold 2004; Jacobs & Thomas 2004). This period is accompanied by syntectonic extensional shearing, late- to post-tectonic intrusions and isothermal decompression (e.g. Jacobs et al. 2003b; Colombo & Talarico 2004). The volume of igneous rocks generated increases towards the end of the extensional period, culminating in voluminous and extensive granitoid magmatism, which is in part charnockitic. Thermobarometry studies show that the charnockites were emplaced as relatively dry melts at temperatures exceeding 900 8C at pressures of c. 5 kbar (Frost & Bucher 1993; Bucher & Frost 1995). Many of the charnockites are retrogressed in part to granite, especially at the contacts and adjacent to late hydrous granite pockets. The geochemistry of the charnockites and associated granitoids is relatively heterogeneous, but they are typically peraluminous to metaluminous and subalkaline with a weak trend to alkaline A-type granites (e.g. Klimov et al. 1964; Ravich & Kamenev 1975; Joshi et al. 1991; Roland 2002, 2004a, b; Li et al. 2003). However, they are not typical A-type granites, being relatively low in Ca, Rb, Nb and Ga, such that they plot as A2-type according to the classificaton of Eby (1992), and unlike common A-type granite associations they form a voluminous and extensive magmatic suite, covering an area of at least 15 000 km2 (Roland 2004a, b). The available geochronology of this late-tectonic stage is summarized in Table 1. As in Dronning Maud Land, the EAAO of northern Mozambique reworks rocks with predominantly Mesoproterozoic protolith ages. However here, the EAAO is divided into two different crustal segments by the ENE –WSW trending Lurio Belt (Figs 1 and 3). The crust south of the Lurio Belt has close similarities to the EAAO of Dronning Maud Land, whereas the EAAO to the north of the Lurio Belt, composed of a collage of terranes, is structurally and lithologically different. The crust south of the Lurio Belt, referred to as the Nampula Complex, is made up of Mesoproterozoic gneisses and migmatites of upper amphibolite facies (e.g. Pinna et al. 1993). This Mesoproterozoic ‘basement complex’ is overlain by: (1) a sequence of Neoproterozoic synorogenic immature clastic sediments, the Mecuburi and Alto Benfica
Groups (e.g. Thomas et al. 2006); (2) tectonic slices (thrust sheets?) of granulite-facies rocks, the Mocuba and Monapo klippen (e.g. Pinna et al. 1993), which might be similar to comparable structures such as the Schirmacher nappe in Dronning Maud Land. The timing of Neoproterozoic – Cambrian amphibolite-facies metamorphism in the Nampula Complex and overlying sedimentary rocks is estimated at 520–490 Ma (e.g. Bingen et al. 2006a, b), whereas the timing of granulitefacies metamorphism in the structurally overlying klippen is dated at 615 + 8 Ma (Kro¨ner et al. 1997). Available data suggest that the Mocuba and Monapo klippen correlate with the Neoproterozoic upper nappes observed north of the Lurio Belt (Pinna et al. 1993). The nature of the observed juxtaposition of the Mesoproterozoic Nampula Complex basement and the granulite klippen is a matter of debate, but previous studies interpreted the contact as a thrust (e.g. Pinna et al. 1993; Kro¨ner et al. 1997). As in Dronning Maud Land, the Nampula Complex is characterized by the presence of large volumes of post-Mesoproterozoic late-tectonic granitoid, with a concentration of such intrusions along the Lurio Belt. Very limited geochronological or geochemical data are available from these rocks. The late-tectonic granite magmatism essentially terminates along the Lurio Belt, although a diminishing number of scattered plutons are found to the north. The Lurio Belt is a northerly inclined high-strain zone, interpreted as a repeatedly reactivated shear zone with an apparent intense late-tectonic pure shear deformation history (Viola et al. 2006). It is marked by isoclinal folding, strongly attenuated lenses of granulites and a characteristic suite of highly sheared leucogneisses. The Lurio Belt shows an intense late-tectonic deformation history as young as c. 500 Ma (Bingen et al. 2006a, b). The belt is an extremely well-defined and coherent structure in the east, prominent in the field and on remote-sensing images, which becomes progressively diffuse and ill-defined towards the west (e.g. Thomas et al. 2006). The EAAO crust north of the Lurio Belt is made up of a collage of generally north–south- to NE – SW-trending nappes. The lower nappes, referred to as the Unango and Marrupa Complexes, are made up of gneisses with Mesoproterozoic protolith ages. These were reworked by high-grade metamorphism dated between c. 560 and 520 Ma (Bingen et al. 2006a, b; Engvik et al. 2007; Norconsult Consortium 2007). The upper nappes, referred to as the Xixano, M’Sawize and Lalamo Complexes, are dominated by rocks with Neoproterozoic protolith ages, and carry evidence for Late Neoproterozoic high-pressure metamorphism dated between 740 and 600 Ma. The upper nappes can
Table 1. Published U–Pb zircon and titanite dates for late- to post-tectonic rocks of Dronning Maud Land, East Antarctica Date (Ma)
Method
Interpretation
Reference
Highly sheared gneiss, Conrad Mts., extensional? Zwiesel gabbro Zwiesel gabbro Late tectonic lamprophyre, Risemedet Charnockitized orthogneiss, Hochlinfjellet Mesoproterozoic metavolcanic rock, Dallmannberge Leucocratic segregation in boudin neck, Conradgebirge Post-tectonic syenite, Humboldtgebirge Hornblende leucosome, Festninga Late-tectonic lamprophyre, Risemedet Migmatite, Jutulsessen Granite, Stabben Aplite dyke in granite, Stabben Post-tectonic granite sheet, Gygra Gabbro, Stabben
530 + 8 527 + 6 521 + 6 523 + 5 521 + 3 522 + 10 516 + 5 512 + 2 510 + 14 508 + 7 504 + 6 500 + 8 495 + 14 487 + 4 483 + 14
U –Pb SHRIMP U –Pb SHRIMP U –Pb SHRIMP U –Pb SHRIMP U –Pb SHRIMP U –Pb SHRIMP U –Pb SHRIMP U –Pb TIMS U –Pb SHRIMP U –Pb SHRIMP U –Pb SIMS U –Pb SIMS Evaporation U –Pb SHRIMP U –Pb SHRIMP*
Metamorphic zircon rim Crystallization Crystallization Crystallization Late charnockitization Metamorphic zircon rim Crystallization of leucosome Crystallization Youngest metamorphic zircon rim Hydrothermal or metasomatic overprint Migmatisation Crystallization Crystallization Crystallization Cooling through 600 8C
Jacobs et al. (1998) Jacobs et al. (2003b) Jacobs et al. (2003b) Jacobs et al. (2003a) Jacobs et al. (2003a) Jacobs et al. (1998) Jacobs et al. (1998) Mikhalsky et al. (1997) Jacobs et al. (2003a) Jacobs et al. (2003a) Paulsson & Austrheim (2003) Paulsson & Austrheim (2003) Paulsson & Austrheim (2003) Jacobs et al. (2003a) Jacobs et al. (2003a)
GRANITES IN EAAO
Lithology and locality
*
Titanite.
73
74
J. JACOBS ET AL.
Fig. 3. Geological overview map of northern Mozambique with sample localities.
possibly be correlated with similar structures in Tanzania and southern Madagascar. The nappe pile north of the Lurio Belt shows a complex and prolonged tectonometamorphic history as a result of an overall NW-directed collisional history along the western margin of the Congo–Tanzania craton between c. 590 and 520 Ma (e.g. Pinna et al. 1993; Jacobs & Thomas 2002; Bingen et al. 2006a, b; Viola et al.
2006). The EAAO north of the Lurio Belt shows few late-tectonic extensional structures or latetectonic igneous rocks compared with the Nampula Complex and Dronning Maud Land. Both in northern Mozambique and Dronning Maud Land, the late-tectonic intrusions have impressive morphological expressions (Fig. 4). In Dronning Maud Land many of the plutons form
GRANITES IN EAAO
Fig. 4. The late-tectonic granitoids form impressive mountains, both in Dronning Maud Land (a) and northern Mozambique (b).
high, steep-sided nunataks, which are not dissimilar in form from the inselberg equivalents in Mozambique.
Analytical details, samples and geochronological results Seven samples were selected for U – Pb secondary ionization mass spectrometry (SIMS) zircon dating, three from Dronning Maud Land and four from NE Mozambique, to constrain the Early Palaeozoic extensional shearing and the late magmatic history of the two areas. From Dronning Maud Land, one sample (J3012/1), containing complex zircons, were collected from a high-strain zone at Armlenet. Two samples (J1670, J1870) from the Mu¨hligHofmann-Gebirge are from different late- to post-tectonic granitoid intrusions at Schneide and Oddesteinen (J1670, J1870) (Fig. 2). In northeastern Mozambique, a sample (JJ238) was selected from a highly foliated granite gneiss sheet, interpreted as syntectonic, another
75
sample (WB295) was selected from a mangerite intrusion within the Lurio Belt (WB295), and two samples (GV01, JJ259) were obtained from two late-tectonic porphyritic granite intrusions with largely undeformed centres and weakly deformed margins (Fig. 3). Zircons were recovered from crushed samples using conventional separation techniques (Wilfley table, magnetic separation, high-density liquids, and hand picking). Zircons were cast, together with a zircon reference standard, in epoxy discs, which were then polished to expose the interiors of the crystals. All zircons were examined and documented in transmitted and reflected light, and cathodoluminescence (CL) images were used to reveal the internal structures of the zircons, such as core –rim relationships in the metamorphic sample (J3012/1). After cleaning, the sample mounts were vacuum-coated with c. 500 nm Au. Zircon U – Pb analysis of five samples were conducted by sensitive high-resolution ion microprobe (SHRIMP) using the SHRIMP II ion microprobe at the John de Laeter Centre of Mass Spectrometry, Curtin University of Technology, Perth, Australia. Two samples were analysed on a Cameca IMS 1270 ion microprobe at the NORDSIM facility, Swedish Museum of Natural History, Stockholm. SHRIMP analytical methods follow those of Williams (1998) and references therein. Analyses consist of six scans through the mass range using a spot size of c. 20 mm diameter. Absolute 238U and 232Th concentrations were determined by comparison with the CZ3 zircon standard (551 ppm 238U) and 238U/206Pb ratios were determined relative to the Temora zircon (417 Ma); standard (206Pb/238U ¼ 0.0668 Black et al. 2003), analyses of which were interspersed with those of unknown zircons. Analyses with the Cameca IMS1270 followed the methods outlined by Whitehouse et al. (1999) and Whitehouse & Kamber (2005), with a spot size of c. 15 mm. The Cameca analyses were calibrated to the Geostandard 91500 reference zircon with an age of 1065 Ma (Wiedenbeck et al. 1995). Uncertainties on U/Pb ratios include propagation of the uncertainty of the day-to-day calibration curve determined from repeated analysis of the reference zircon. A common Pb correction was performed using non-radiogenic 204Pb with an average modern crustal Pb composition (Stacey & Kramers 1975). Data were analysed using software programs SQUID and ISOPLOT (Ludwig 2001, 2003). Uncertainties in ratios and ages in Table 2 are listed at the 1s level. Weighted mean ages are quoted below with 95% uncertainties.
76
Table 2. Ion microprobe data on zircon from Dronning Maud Land and Mozambique Identifier 1
U (ppm) 2
3
Th (ppm)
Pb (ppm)
f (%) 4
238
U/ Pb 5
206
1s (%)
207
Pb/ Pb 5
206
1s (%)
238
U/ Pb 6
206
1s (%)
207
Pb/ Pb 6
1s (%)
206
206 238
Pb/ U (Ma) 7
1s
207
Pb/ Pb (Ma) 8
206
1s
Disc (%) 9
2.2 2.3 2.1 2.2 2.2 2.1 2.4 2.2 2.4 2.1 5.8
0.0575 0.0516 0.0568 0.0574 0.0563 0.0570 0.0757 0.0751 0.0770 0.0761 0.0769
1.7 4.2 0.8 2.1 1.3 0.7 0.9 0.6 1.6 0.8 2.4
498 507 508 514 518 522 972 1087 1128 1147 1166
10 11 10 11 11 11 22 23 26 24 65
513 268 483 507 464 490 1086 1070 1121 1098 1146
38 96 18 45 28 16 17 12 32 17 47
3 283 25 21 211 26 10 21 21 24 22
J1670, post-tectonic granite, Dronning Maud Land, Oddesteinen, SHRIMP Perth 4.1 106 83 0.05 12.82 1.3 0.0577 2.0 5.2 59 40 0.38 12.67 1.4 0.0613 2.6 6.1 90 102 0.11 12.61 2.3 0.0574 2.2 12.1 236 179 0.17 12.53 1.3 0.0571 1.8 10.1 212 67 0.15 12.52 1.2 0.0580 1.4 5.1 94 51 0.00 12.46 1.6 0.0597 2.1 11.1 183 140 0.00 12.42 1.2 0.0557 1.6 3.1 117 76 0.29 12.32 1.3 0.0575 1.9 1.1 287 146 0.00 12.33 1.2 0.0570 1.3 9.1 259 146 0.18 12.30 1.3 0.0571 1.3 2.1 196 109 0.07 12.30 1.5 0.0577 1.5 7.1 173 71 0.00 12.29 1.4 0.0593 1.6 9.2 336 130 0.15 12.14 1.8 0.0572 1.6 8.1 157 53 0.16 12.09 1.6 0.0576 1.9
12.83 12.72 12.62 12.55 12.53 12.46 12.42 12.36 12.33 12.32 12.31 12.29 12.16 12.11
1.3 1.5 2.3 1.3 1.2 1.6 1.2 1.3 1.2 1.3 1.5 1.4 1.8 1.6
0.0573 0.0583 0.0566 0.0557 0.0567 0.0597 0.0557 0.0552 0.0570 0.0556 0.0571 0.0593 0.0560 0.0563
3.7 5.4 4.3 2.7 2.4 2.1 1.6 3.7 1.3 1.7 1.8 1.6 2.6 2.7
484 487 492 495 495 496 500 503 503 504 504 503 510 512
6 7 11 6 6 8 6 6 6 6 7 7 9 8
503 539 474 440 480 657 486 419 515 438 496 604 453 463
82 119 96 61 53 55 40 83 32 38 40 37 57 60
4 10 23 212 23 24 23 219 2 215 21 16 212 210
J1870, post-tectonic granite, Dronning Maud Land, Schneide, 1.1 69 42 0.36 12.56 2.1 170 105 0.00 12.53 4.1 166 119 0.00 12.50 10.1 99 65 0.81 12.35 9.1 76 52 0.66 12.28 7.1 205 118 0.16 12.28
12.60 12.53 12.50 12.45 12.36 12.30
2.2 2.0 2.1 2.1 2.2 1.9
0.0545 0.0567 0.0586 0.0525 0.0536 0.0572
7.8 1.9 1.9 4.8 6.5 1.9
494 495 495 501 504 504
11 10 10 10 11 10
390 612 580 307 356 500
176 76 44 109 146 43
226 19 14 262 241 21
SHRIMP Perth 2.2 0.0574 2.0 0.0567 2.1 0.0586 2.1 0.0591 2.1 0.0590 1.9 0.0586
2.9 1.9 1.9 2.3 2.7 1.8
J. JACOBS ET AL.
J3012, sheared felsic gneiss, Dronning Maud Land, Armlenet, Gjelsvikfjella, SHRIMP Perth 9.1 748 64 0.15 12.43 2.2 0.0587 1.3 12.45 10.1 x 607 34 0.91 12.19 2.2 0.0590 1.6 12.30 11.1 2107 9 0.03 12.20 2.1 0.0570 0.7 12.20 6.1 331 44 0.07 12.04 2.2 0.0579 1.6 12.05 4.1 1364 1 0.09 11.96 2.2 0.0571 0.9 11.97 5.1 1625 3 0.01 11.88 2.1 0.0570 0.7 11.88 8.1 x 700 37 0.13 6.11 2.4 0.0767 0.7 6.11 3.1 1954 48 0.09 5.44 2.2 0.0758 0.6 5.45 7.1 570 148 0.02 5.23 2.4 0.0772 1.6 5.23 1.1 452 145 0.07 5.14 2.1 0.0767 0.8 5.15 2.1 269 84 0.00 5.05 5.8 0.0769 2.4 5.05
173 244 440 69
2.3 1.9 1.9 2.2
0.0556 0.0549 0.0549 0.0577
2.9 2.3 1.7 2.9
506 508 509 513
11 10 9 11
438 410 408 541
64 51 38 65
215 223 224 5
JJ238, granite gneiss, 11.1 x 10.1 x 1.1 14.1 12.1 16.1 9.1 13.1 4.1 2.1 3.1 7.1 8.1 15.1 6.1 5.1
Mozambique, Nampula Complex, UTM: Zone37, 266082, 8299521, SHRIMP 189 240 0.12 12.97 1.1 0.0588 1.2 12.98 1.1 277 75 0.08 12.75 0.9 0.0574 0.9 12.76 0.9 144 179 0.00 12.43 1.1 0.0571 1.4 12.43 1.1 199 230 0.17 12.40 1.0 0.0577 1.2 12.42 1.0 319 104 0.00 12.29 0.9 0.0563 0.9 12.29 0.9 274 214 0.10 12.27 1.0 0.0569 1.0 12.29 1.0 275 115 0.01 12.27 1.0 0.0564 1.0 12.27 1.0 412 1293 0.00 12.25 0.9 0.0574 0.8 12.25 0.9 270 172 0.00 12.21 1.0 0.0578 1.0 12.21 1.0 298 99 0.19 12.16 1.0 0.0575 1.0 12.18 1.0 311 97 0.16 12.14 1.0 0.0577 0.9 12.16 1.0 230 84 0.12 12.14 1.0 0.0566 1.1 12.15 1.0 263 736 0.10 12.10 1.0 0.0572 1.0 12.11 1.0 243 379 0.20 12.08 1.0 0.0594 1.1 12.10 1.0 269 182 0.23 12.07 1.0 0.0577 1.0 12.09 1.0 262 674 0.00 11.98 1.0 0.0579 1.1 11.98 1.0
Perth 0.0579 0.0568 0.0571 0.0563 0.0563 0.0561 0.0563 0.0574 0.0578 0.0559 0.0564 0.0556 0.0564 0.0578 0.0558 0.0579
1.8 1.3 1.4 1.8 0.9 1.7 1.1 0.8 1.0 1.6 1.8 1.4 1.6 1.9 1.8 1.1
478 486 499 500 505 505 506 506 507 510 510 511 512 512 513 517
5 4 5 5 5 5 5 5 5 5 5 5 5 5 5 5
525 484 551 464 499 456 464 514 524 450 470 435 469 522 446 546
39 29 40 39 29 37 25 18 22 35 39 32 34 43 41 25
10 21 10 27 21 210 28 2 3 211 28 215 28 2 213 6
498 494 503 504 505 507 512 513 515 516 517 517 518 519 563
8 8 8 8 8 8 8 8 8 8 8 8 8 8 9
201 590 474 491 481 467 488 485 512 491 452 494 477 489 587
133 56 15 39 24 19 16 24 11 15 29 14 20 18 28
2151 17 26 23 25 29 25 26 21 25 215 25 29 26 4
105 192 186 36
0.17 0.33 0.22 0.00
12.25 12.20 12.18 12.08
2.3 1.9 1.9 2.2
0.0570 0.0576 0.0567 0.0577
1.8 1.6 1.1 2.9
12.27 12.24 12.21 12.08
JJ259, Phenocryst granite, Mozambique, Nampula Complex, UTM: Zone 37, 270036, 8290067, CAMECA Stockholm 01b c x 115 257 15 1.00 12.44 1.5 0.0580 3.0 12.57 1.5 0.0501 6.0 16b c x 68 90 8 0.69 12.50 1.6 0.0596 2.6 12.50 1.6 0.0596 2.6 17a r 1187 206 106 0.08 12.31 1.5 0.0571 0.7 12.32 1.5 0.0565 0.7 01a r 1553 885 153 1.07 12.16 1.5 0.0653 0.9 12.30 1.5 0.0570 1.8 02a c 1391 320 126 0.10 12.26 1.5 0.0575 1.0 12.28 1.5 0.0567 1.1 13a r 808 197 74 0.06 12.23 1.5 0.0568 0.8 12.24 1.5 0.0564 0.8 17b r 1311 206 119 0.12 12.12 1.5 0.0569 0.7 12.12 1.5 0.0569 0.7 05a c 1024 147 92 0.13 12.07 1.5 0.0578 1.0 12.08 1.5 0.0568 1.1 10b r 1174 95 106 0.02 12.02 1.6 0.0577 0.5 12.02 1.6 0.0575 0.5 10a c 725 57 65 0.07 12.00 1.6 0.0576 0.6 12.01 1.6 0.0570 0.7 06a r 1036 86 93 0.32 11.97 1.6 0.0585 1.0 12.01 1.6 0.0560 1.3 16a r 979 75 88 0.08 11.98 1.5 0.0577 0.6 11.99 1.5 0.0571 0.7 08a c 449 539 53 0.07 11.97 1.6 0.0571 0.9 11.97 1.6 0.0566 0.9 04a r 1716 261 157 0.15 11.95 1.6 0.0569 0.8 11.95 1.6 0.0569 0.8 13b c x 186 50 19 0.15 10.94 1.6 0.0596 1.3 10.94 1.6 0.0596 1.3
GRANITES IN EAAO
5.1 3.1 6.1 8.1
(Continued) 77
78
Table 2. Continued Identifier 1
U (ppm) 2
3
Th (ppm)
Pb (ppm)
f (%) 4
238
U/ Pb 5
206
1s (%)
207
Pb/ Pb 5
206
1s (%)
238
U/ Pb 6
206
1s (%)
207
Pb/ Pb 6
206
1s (%)
Pb/ U (Ma) 7
500 500 501 504 505 505 506 507 507 508 508 508 508 508 508 510 512 511 513 514 514 515 515 517 525 531 625
1s
5 5 6 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 6
207 206
Pb/ Pb (Ma) 8
506 514 546 501 488 500 513 488 512 503 512 488 510 511 517 508 494 528 491 497 520 499 516 526 281 486 704
1s
Disc (%) 9
17 11 30 18 21 15 15 13 13 11 19 11 15 16 13 15 10 23 12 11 18 15 13 19 20 12 11
1 3 8 21 23 21 1 24 1 21 1 24 0 0 2 21 24 3 25 23 1 23 0 2 289 29 11
J. JACOBS ET AL.
GV01, Phenocryst granite, Mozambique, Nampula Complex, UTM: Zone 37, 426538, 8351026, CAMECA Stockholm 48 c 600 440 62 0.08 12.40 1.1 0.0580 0.8 12.41 1.1 0.0574 0.8 47 c 1018 315 96 0.10 12.38 1.1 0.0583 0.5 12.39 1.1 0.0576 0.5 61 c 420 286 46 1.00 12.22 1.2 0.0662 0.8 12.35 1.3 0.0584 1.4 64 r 722 125 65 0.15 12.28 0.9 0.0584 0.7 12.30 0.9 0.0572 0.8 49 c 492 203 48 0.04 12.28 1.0 0.0572 0.9 12.28 1.0 0.0569 0.9 61 r 694 159 64 0.05 12.27 1.0 0.0576 0.7 12.28 1.0 0.0572 0.7 50 c 487 834 63 0.07 12.22 1.0 0.0581 0.6 12.23 1.0 0.0576 0.7 59 r 772 179 71 0.05 12.23 0.9 0.0573 0.6 12.23 0.9 0.0569 0.6 62 r 670 124 61 0.06 12.21 1.0 0.0580 0.6 12.22 1.0 0.0575 0.6 52 r 1018 84 90 0.06 12.19 1.1 0.0578 0.5 12.20 1.1 0.0573 0.5 51 c 347 117 33 0.16 12.18 1.0 0.0587 0.8 12.20 1.0 0.0575 0.9 53 r 944 90 84 0.02 12.20 1.0 0.0571 0.5 12.20 1.0 0.0569 0.5 57 c 530 196 51 0.05 12.19 0.9 0.0578 0.7 12.20 0.9 0.0575 0.7 56 c 440 463 50 0.05 12.18 0.9 0.0579 0.7 12.19 0.9 0.0575 0.7 45 c 578 336 59 0.02 12.18 1.0 0.0578 0.6 12.18 1.0 0.0577 0.6 52 c 535 269 54 0.10 12.13 1.0 0.0582 0.6 12.14 1.0 0.0574 0.7 57.1 c 1274 432 121 0.05 12.10 0.9 0.0575 0.4 12.11 0.9 0.0571 0.4 54 c 528 875 68 0.14 12.09 0.9 0.0591 1.0 12.10 0.9 0.0580 1.0 60 r 826 164 77 0.04 12.09 1.0 0.0573 0.5 12.09 1.0 0.0570 0.6 45 r 769 76 69 0.02 12.06 1.0 0.0573 0.5 12.06 1.0 0.0571 0.5 57.2 c 619 327 62 0.04 12.04 0.9 0.0581 0.8 12.05 0.9 0.0577 0.8 46 c 475 167 46 0.03 12.02 1.0 0.0574 0.7 12.03 1.0 0.0572 0.7 63 r 753 98 69 0.05 12.02 0.9 0.0581 0.5 12.02 0.9 0.0576 0.6 63 c 380 289 41 0.18 11.96 1.1 0.0593 0.8 11.98 1.1 0.0579 0.9 58 r x 959 171 89 0.71 11.79 0.9 0.0575 0.5 11.88 0.9 0.0519 0.9 65 c x 933 310 93 0.08 11.66 1.0 0.0575 0.5 11.67 1.0 0.0569 0.5 55 c x 1544 887 192 0.06 9.78 0.9 0.0633 0.5 9.79 0.9 0.0629 0.5
206 238
422 476 501 505 505 512 507 509 503 511 514 519 510 518 514 521 523 528
5 16 7 8 7 10 5 9 8 7 6 9 8 6 8 8 8 13
270 972 399 585 616 178 544 518 906 452 587 307 884 464 833 688 775 490
87 238 200 113 120 307 56 168 237 101 62 286 138 52 170 114 122 208
236 100 220 16 22 265 7 2 78 211 14 241 71 210 60 31 47 27
1, analysis identifier. 2, r, rim; c, core. 3, x, not selected for age calculation. 4, per cent of total 206Pb made by common Pb. 5, ratios uncorrected for common Pb. 6, ratios corrected for common Pb using 204Pb method, if correction is positive. 7, age corrected for common Pb with the 207Pb method. 8, age corrected for common Pb with the 204Pb method. 9, discordance of the analysis; positive number for an analysis above the concordia curve in a Tera– Wasserburg diagram.
GRANITES IN EAAO
WB295, Quartz mangerite, Mozambique, Unango Complex, UTM: Zone 37, 219462, 8288506, SHRIMP Perth 8.1 x 79 26 0.20 14.83 1.2 0.0533 2.0 14.86 1.2 0.0516 3.8 6.1 x 26 22 0.00 13.02 3.5 0.0580 4.0 13.02 3.5 0.0580 4.0 4.1 45 39 0.26 12.38 1.5 0.0568 2.4 12.41 1.6 0.0547 8.9 11.1 38 34 0.17 12.23 1.5 0.0608 2.5 12.25 1.6 0.0595 5.2 1.1 28 16 0.00 12.26 1.4 0.0585 3.0 12.26 1.4 0.0585 3.0 8.1 20 18 1.14 12.09 1.9 0.0590 3.4 12.22 2.0 0.0496 13.2 13.1 101 85 0.14 12.19 1.1 0.0595 1.4 12.20 1.1 0.0584 2.6 9.1 24 23 0.03 12.16 1.8 0.0579 3.1 12.17 1.8 0.0577 7.7 17.1 26 25 0.00 12.30 1.7 0.0590 2.9 12.30 1.7 0.0590 2.9 18.1 37 17 0.28 12.11 1.5 0.0583 2.4 12.14 1.5 0.0560 4.5 14.1 56 46 0.00 12.05 1.3 0.0580 2.0 12.05 1.3 0.0580 2.0 15.1 24 10 0.64 11.92 1.7 0.0578 3.1 12.00 1.8 0.0525 12.5 7.1 28 13 0.00 12.08 1.7 0.0623 2.9 12.08 1.7 0.0623 2.9 12.1 90 67 0.17 11.96 1.1 0.0577 1.5 11.98 1.1 0.0563 2.4 3.1 33 21 0.00 12.04 1.6 0.0583 2.7 12.04 1.6 0.0583 2.7 10.1 35 20 0.00 11.89 1.5 0.0574 2.6 11.89 1.5 0.0574 2.6 16.1 30 25 0.00 11.83 1.6 0.0579 2.7 11.83 1.6 0.0579 2.7 2.1 31 30 0.46 11.67 2.5 0.0608 2.7 11.72 2.6 0.0570 9.4
79
80
J. JACOBS ET AL.
Results Sample J3012/1, mylonitic felsic gneiss, Armlenet, Dronning Maud Land Sample J3012/1 is a mylonitic felsic gneiss from Armlenet and was collected from the central part of a high-strain zone (Fig. 5), several tens of metres wide, which affected mainly granitic gneisses and augen gneisses as well as highly dismembered amphibolites. The shear zone is cut by pegmatite veins (Fig. 5d) that are in places also sheared. The central part of the shear zone shows high shear strain with no mesoscopic shear-sense indicators, although it contains oblique folds with fold axes dipping towards the NNE. Closer to the shear zone margins at lower shear strains, abundant shear-sense indicators signify extension, with NNEplunging stretching lineations. The analysed sample is equigranular and consists of microcline, partly antiperthitic plagioclase, quartz, and dark brown biotite. Both straight and bulged grain boundaries of quartz are common. This sample contains small (50 –150 mm), clear and colourless to dark brown, equant to elongate zircons, with aspect ratios up to 5:1. Many zircons
are metamict and various inclusions are common. Many show zircon overgrowth and/or resorption features, and several show necking. In CL, the complex internal structure of the zircons becomes apparent (Fig. 6), with some grains showing oscillatory concentric growth zoning in their cores. These cores have a high-U reaction zone of varying width. It is often irregular and appears to be a diffusion front, rather than a simple overgrowth. Eleven areas in five cores and six rims were analysed (Fig. 7a and b). The zircon cores contain 269– 2154 ppm 238U and have Th/U of 0.03– 0.33, whereas rims have 331– 2107 ppm 238U and Th/U from almost zero (0.0006) to 0.14 (Table 2). The proportion (f204) of common 206Pb to total 206Pb is low (,1% for all analyses). The five core analyses form a concordant to slightly discordant group (Fig. 7a) with a mean 207Pb/206Pb age of 1086 + 25 Ma (MSWD ¼ 1.2). Four of them yield a concordia age (Ludwig 1998) of 1098 + 25 Ma (MSWD ¼ 1.9). Five of the six rim analyses plot close to concordia (Fig. 7b). The rim that produced reversely discordant analysis 10.1 has relatively low U and the largest common Pb correction, and its 207Pb/206Pb ratio is probably over-corrected for common Pb. However, this has a
Fig. 5. High-strain extensional shear zone at Armlenet. (a, b) Highly sheared grey migmatic gneisses and sample locality J3012/2; (c) complex folded part of the shear zone; (d) mylonitic gneisses intruded by felsic leucosomes.
GRANITES IN EAAO
81
Fig. 6. CL image of selected zircons from mylonite sample J3012/2. Many of the grains are highly complex and show strong recrystallization, probably as the result of high-temperature shearing.
negligible effect on its 238U/206Pb ratio and the weighted mean 238U/206Pb age for all six rim analysis is 507 + 9 Ma (MSWD ¼ 0.99). The age of 1098 + 25 Ma for zircon cores is interpreted as the age of crystallization of the protolith of the gneiss, whereas the age of 507 + 9 Ma for six rim analyses is interpreted to reflect the time of metamorphism during shearing along the Armlenet Shear Zone.
Sample J1670, granite, Oddesteinen, Dronning Maud Land Sample J1670 is a coarse-grained granitoid from Oddesteinen. It is composed of approximately equal amounts of plagioclase, perthitic K-feldspar, quartz, and olive –green hornblende. Accessory minerals include xenomorphic titanite, biotite, apatite and opaque minerals. The sample contains clear and colourless to pale yellow zircons, which commonly occur as inclusions in hornblende. Most are fragments of elongate crystals up to 300 mm in length, with aspect ratios up to 5:1. Some crystals have rounded terminations. Some contain large vermicular (worm-like) melt inclusions, and evidence of resorption is also common. CL images reveal oscillatory growth zoning, but also irregular zoning and recrystallization (Fig. 7c). Some zircon cores are high in U and metamict. Fourteen areas in 13 zircons were analysed, mostly targeting material showing magmatic oscillatory growth zoning (Fig. 7c). The analysed areas
have low common Pb ( f204 ,0.29), and low to moderate 238U concentrations of 59–336 ppm and Th/U of 0.33–1.17, values typical for magmatic zircon. The data form a concordant to very slightly discordant group, and all 14 analyses yield a concordia age of 499 + 4 Ma (MSWD ¼ 1.2). This result is interpreted as the crystallization age of the Oddesteinen granite.
Sample J1870, post-tectonic charnockite, Schneide, Dronning Maud Land Sample J1870 was collected from a post-tectonic charnockitic granitoid at Schneide. The sample consists of perthitic K-feldspar, plagioclase, minor quartz, and relics of altered ortho- and clinopyroxene and hornblende. K-feldspar contains large patchy mesoperthite exsolutions of plagioclase and is typically surrounded by myrmekite. Orthopyroxene has a reaction zone consisting of hornblende, plagioclase and opaque minerals. Apatite and allanite are accessory phases. The sample contains colourless to brown, and equant to acicular (aspect ratios up to 6:1) zircons and zircon fragments. Some zircons possess high-U cores and many contain abundant inclusions. Several zircons have long prismparallel cracks. CL images reveal strong oscillatory growth zoning (Fig. 7d), which was targeted for analysis; high-U cores were avoided. The analysed areas contain 69 – 440 ppm 238U and have Th/U of 0.44 – 0.81. Common Pb is
82
J. JACOBS ET AL.
Fig. 7. U–Pb analytical results for samples: (a, b) J3012/1; (c) J1670; (d) J1870; (e) JJ238; (f) JJ259; (g) GV01; (h) WB295.
GRANITES IN EAAO
low ( f204 ,0.8%). Ten areas from 10 zircons were analysed (Table 2). The measured compositions are concordant to slightly discordant and yield a concordia age of 501+ 7 Ma (MSWD ¼ 1.12), which is interpreted as the crystallization age of this charnockitic granite.
Sample WB295, mangerite, Lurio Belt, northern Mozambique WB295 was sampled from a mangerite intrusion immediately to the north of the Lurio Belt within the Unango Complex (Fig. 3). The sample contains large (up to 400 mm), stubby (length –width ratios ,2), clear, colourless to light brown zircons, many of which contain large inclusions of other minerals. Although some are euhedral, most zircons are anhedral and irregular in shape. In CL, irregular and sector zoning is weakly to moderately defined in most zircons. Some exhibit weak oscillatory growth zoning, and several zircons show reaction zones. Eighteen areas were analysed in 18 zircons (Fig. 7h). All analyses are low in U (20–100 ppm) and Th (10–85 ppm) and have Th/U between 0.3 and 1, typical of magmatic zircon. Common Pb ranges up to 1.1%. Excluding two significantly younger analyses (8.1 and 6.1), 16 analyses form a mainly concordant group with a concordia age of 512 + 4 Ma (MSWD ¼ 1.2), which is interpreted as the crystallization age of the mangerite.
Sample JJ238, granite gneiss, Gurue`, northern Mozambique Sample JJ238 is a medium- to coarse-grained, highly foliated biotite granite gneiss collected close to Gurue in the Nampula Complex, not far from the boundary with the Lurio Belt. The sample contains trace amounts of titanite, allanite, apatite and opaque minerals. Zircon is abundant and occurs as a simple population of clear and elongate crystals, up to 200 mm long and with length–width ratios of up to six. Several have rounded terminations, and many contain mineral inclusions. In CL, the cores of most zircons exhibit concentric growth zoning, and are surrounded by thin reaction zones and overgrown by thin zircon rims. Some cores show evidence of resorption and alteration prior to overgrowth by zircon rims. The oscillatory zoned parts of 16 different zircons were analysed; the rims were not sufficiently wide to be analysed (Fig. 7e). Uranium concentration ranges from c. 140 to 410 ppm, and Th/U from 0.3 and 3.2, typical for igneous zircons. The common lead is uniformly low (see value of f204 in Table 2 with a mean of 0.07%). Fourteen
83
analyses form a concordant group with a concordia age of 507 + 3 Ma (MSWD ¼ 1.4), which we interpret as the crystallization age of the protolith of the biotite gneiss. Two slightly younger analyses (10.1 and 11.1) may reflect minor Pb loss from the analysed areas, possibly at the time the thin zircon rims were formed, during deformation of the gneiss.
Sample JJ259, porphyritic biotite–hornblende granite, Gurue`, northern Mozambique Sample JJ259 was collected from a small (a few kilometres long) elongate porphyritic granite pluton in the Nampula Complex. K-feldspar phenocrysts are up to 6 cm in length and locally show a well-defined igneous flow foliation. The sample contains biotite and amphibole and trace amounts of allanite and apatite. Zircons are equant to elongate and are characterized in CL images by an oscillatory-zoned core and a U-rich rim, commonly growing at the expense of the core. Fifteen analyses were made in 10 crystals. Uranium concentration ranges from 68 to 1716 ppm, and Th/U from 0.08 and 2.2. Common lead is generally low, but f204 is up to 1% for two analyses. No age difference can be detected between cores and rims. Twelve analyses define a concordia age of 508 + 4 Ma (Fig. 7f), interpreted as the intrusion age of the granite. One analysis (13b) in a core is significantly older (563 + 18 Ma), suggesting the presence of inherited material in the cores of the zircons. Two analyses in low-U cores are dispersed from the main group owing to the presence of significant common Pb.
Sample GV01, porphyritic biotite–hornblende granite, Riba´ue`, northern Mozambique Sample GV01 is a porphyritic granite, collected about 2 km north of Riba´ue` in the marginal part of a major, at least 60 km long, late-tectonic granite intrusion, roughly aligned with the regional ENE –WSW regional fabric. At the sampling locality, a planar fabric is well defined by Kfeldspar phenocrysts (1– 3 cm in length), possibly produced by magmatic flow or solid-state deformation. The sample contains biotite, and minor amounts of amphibole, muscovite, titanite, allanite, apatite and opaque minerals. Zircon forms elongate, transparent to light brown, zoned crystals up to 300 mm long, with aspect ratios up to 5:1. The zircons have irregular outlines, rounded terminations, and abundant mineral inclusions. CL images reveal core –rim structures, with oscillatory-zoned cores and rims up to 30 mm wide. The rims might have grown during late intrusive deformation.
84
J. JACOBS ET AL.
Twenty-seven areas were analysed in 21 grains, including 10 analyses of rims and 17 of cores (Fig. 7g). The rims contain 670 –1018 ppm 238U, have Th/U of 0.08 –0.23, and low common Pb ( f204 ,0.7%). The cores contain 347– 1544 ppm 238 U, have Th/U of 0.3–1.7, and have mainly low common Pb (f204 ,0.18% for 16 of 17 analyses). The cores and rims are indistinguishable in age and combine to yield a concordia age of 508 + 2 Ma. This age is interpreted as bracketing the crystallization age of the granite and the late-intrusive deformation around its margin.
Geochemical characteristics and tectonic interpretation of the late- to post-tectonic magmatic province in Dronning Maud Land The late-tectonic magmatism in Dronning Maud Land occurred between c. 530 and 485 Ma. It began with subordinate gabbro intrusions at c. 530– 520 Ma (e.g. Jacobs et al. 2003a) and terminated with similarly small amounts of sheet-like granite intrusions at c. 485 Ma (e.g. Jacobs et al. 2003b). Between these events, the major pulse of igneous activity was relatively short-lived between c. 510 and 500 Ma. The two samples dated in this study, the Oddesteinen granite dated at 499 + 4 Ma and the Schneide charnockite at 501+ 7 Ma, were emplaced towards the end of this interval and consequently confirm this previously recognized trend. Probably more than 90% of the latetectonic igneous rocks were intruded during this time interval. Fewer published data are available on the Nampula Complex in NE Mozambique. The four samples dated in this study represent four different occurrences of late-orogenic granitoids; a mangerite pluton (WB295), a sheared granite orthogneiss (JJ238), a porphyritic granite sheet (JJ259) and a large, weakly deformed porphyritic granite pluton (GV01). The samples gave dates of 512 + 4, 507 + 3, 508 + 4 and 508 + 2 Ma, respectively, indicating that the main pulse of late-orogenic magmatism in the Nampula Complex and along the Lurio Belt was synchronous with, or slightly older than, in Dronning Maud Land. In Dronning Maud Land, the main magmatism is associated with coarse-grained to megacrystic, generally undeformed, granitoid –charnockite bodies, that are exposed over an estimated area of more than 15 000 km2 between 28E and 288E in central and eastern Dronning Maud Land (e.g. van Autenboer & Loy 1972; Shiraishi et al. 1983, 1994; Ohta et al. 1990; Moyes et al. 1993; Paech 2001; Owada et al. 2003; Paech et al. 2004; Roland 2004a, b; Bisnath et al. 2006). A compositional
range from granite, quartz monzonite, monzonite, syenite and minor anorthosite is recognized, typically associated with reddish brown weathering charnockites, which are volumetrically the most abundant rocks exposed. The charnockites are composed of quartz, mesoperthite and plagioclase, with (especially in magnesian varieties) primary anhydrous mafic phases of pigeonite and augite. Ironrich varieties are typified by coexisting fayalite þ hedenbergite þ quartz (Frost & Bucher 1993). The original mafic phases are very often replaced or overgrown by hornblende and biotite, indicating hydration by late-magmatic fluids. Thermobarometrical studies by Frost & Bucher (1993) and Bucher & Frost (1995), and petrological studies by Markl & Henjes-Kunst (2004), suggest that the charnockites were emplaced as relatively dry melts at temperatures exceeding 900 8C and pressures of c. 4.8 kbar. When the charnockites cooled to below 800 8C, the remaining melt pockets became water saturated, causing hydration of the mineralogy. Furthermore, inclusions of gneiss and adjacent host rocks expelled fluids to flux the remaining melt and led to the crystallization of mafic hydrous silicates. These late hornblende– biotite granitic melts expelled aqueous fluids during their crystallization, which invaded adjacent parts of charnockite plutons and altered them to biotite–hornblende granodiorite. Such processes produced bleached zones in the otherwise brownish charnockites. Several papers on the bulk geochemical characteristics of the charnockites in Dronning Maud Land and their retrogressed varieties have been published (e.g. Klimov et al. 1964; Ravich & Soloviev 1966; Joshi et al. 1991; Roland 2002, 2004a, b; Jacobs et al. 2003a, b; Li et al. 2003; Paulsson & Austrheim 2003; Bauer & Jacobs 2005; Engvik et al. 2005), but the interpretation of the results and the implications for the geotectonic setting remain ambiguous. Geochemically, the rocks can be characterized as peraluminous to metaluminous, or mildly alkaline granites. In contrast to the contemporaneous, Mg-rich, charnockites of India and Sri Lanka the rocks from Dronning Maud Land are very enriched in Fe, manifested as coexisting ferrosilite and/or fayalite and quartz. Most of the Dronning Maud Land charnockites and related granitoids are enriched in Ba, Sr, Zr, Y, Zn and Fe, but they are relatively low in Cs, Rb and Ca normalized to a primitive mantle composition (e.g. Markl & Henjes-Kunst 2004; Roland 2004a, b). That is, they represent highly fractionated melts from a mainly mafic to intermediate source but are not typical A-type granitoids as originally defined by Loiselle & Wones (1979). Nevertheless, on various discrimination diagrams, the rocks broadly plot into the ‘within-plate’ tectonic setting field of the granitoids, although such
GRANITES IN EAAO
85
diagrams were defined for A-type granitoids, which represent differentiates of magmas derived from mantle material (Eby 1992). Eby (1992) defined this subtype to differentiate intra-plate rift-related (A1-type) granitoids from post-collisional (A2type) granitoids. The geochemical signature of the charnockites and related granitoids in central Dronning Maud Land coincides with the field of A2-type granitoids (Fig. 8), which represent magmas derived from continental crust of tonalitic to granodioritic composition or underplated crust. According to Roland (2002, 2004a, b) the charnockites were derived from lower continental or underplated crust; this derivation would explain their relatively heterogeneous geochemical signature. The megacrystic, sometimes rapakivi-type textures often seen in such rocks was recently interpreted by Bonin (2007) as an indicator for an emplacement at a middle crustal level near the ductile –brittle transition zone (3–5 kbar). Partial melting of a mafic source of charnockitic magma requires supply of excess heat in a thickened lithosphere (Bonin 2007). The extensive high-temperature dry melt generation in the lower crust could have been accomplished by continental lithospheric mantle delamination after the collision between parts of East and West Gondwana, with the heat source provided by the large-scale uprise of hot asthenosphere (Fig. 9). The resulting highly Fig. 9. Delamination model of the southern part of the EAAO can explain the structurally defined extent of large volumes of high-temperature, A2-type granitoids within the southern part of the EAAO. (a) Collision of part of East and West Gondwana; (b) delamination of the orogen root and influx of asthenosphere (arrows); (c) generation of voluminous melts and orogenic collapse of the orogen. Cross-section is approximately east–west across the Nampula Province.
Fig. 8. Composition of charnockites and related granitoids from Dronning Maud Land (DML) plotted in the Y– Nb– 3Ga diagram after Eby (1992). Four analyses of samples from Mozambique are plotted as filled triangles.
fractionated charnockitic magmas intruded the high-grade metamorphic basement at c. 500 Ma and cooled to below c. 300 8C only by around 460–450 Ma giving a relatively slow cooling rate of 10 8C Ma21 (Markl & Henjes-Kunst 2004). This slow cooling rate was presumably due to the high volumes of hot melt invading the lower to middle crust. The spatial distribution of the granitoid– charnockite plutons appears to be structurally controlled (i.e. within the boundaries of the southern third of the EAAO), effectively terminating against the Lurio Belt in the north. Markl & Henjes-Kunst (2004) recorded significant variation in the Nd isotopic composition of the post-tectonic plutons of Dronning Maud Land, reflecting differing crustal components in their parental melts. For example, 1Nd values of around 23
86
J. JACOBS ET AL.
in the eastern Mu¨hlig-Hofmann-Gebirge (Fig. 1) indicate a significant contribution from Mesoproterozoic crust in this area, whereas values of around 210 in the Orvinfjella suggest that Paleoproterozoic crustal components were present. Further isotopic studies could thus be used to test and map the age of protoliths in Dronning Maud Land. For example, the Mesoproterozoic orogenic belt fringing the Kaapvaal–Grunehogna craton possibly terminates between the eastern Mu¨hlig-Hofmann-Gebirge and Orvinfjella (Bauer et al. 2003).
Tectonic models How can the widespread late-tectonic magmatic event described here be accommodated within the current tectonic models for the southern part of the EAAO? (1) The early French–Mozambiquan work in northern Mozambique, summarized by Pinna et al. (1993), interpreted the thrust-bound packages to the north of the Lurio Belt as the remains of a major latest Mesoproterozoic–Early Neoproterozoic (1100–950 Ma) thrust pile that was overthrust SE along the Lurio Belt over the Nampula Province. In this model Late Neoproterozoic and Early Palaeozoic isotopic ages from various rocks were assigned to a mainly thermal event with little tectonic component. Clearly, from the recent new work and the plentiful Palaeozoic metamorphic ages from the Lurio Belt, this model is now untenable. (2) A similar tectonic model but with different timing has been suggested by Grantham et al. (2003, 2007), who inferred SE-directed thrusting of granulite-facies rocks along the Lurio Belt over the Nampula Complex (their ‘Nampula Subprovince’) to the south between c. 620 and 550 Ma. In this model, the Monapo and Mugeba klippen are interpreted as the exposed remains of this once more widespread nappe complex. Grantham et al. furthermore suggested that Late Neoproterozoic nappes north of the Lurio Belt are part of a much larger nappe system, which extended into Sri Lanka and southern Madagascar to the east, and possibly to the Damara Belt of Namibia to the west. The Lurio Belt is interpreted as a major suture in this model. Grantham et al. (2007) suggested that the late granitoid magmatism in the Nampula Complex resulted from loading by the major thrust stack and it is argued that the time gap of c. 50 Ma between thrusting and granitoid intrusions is the normal delay time for the crust to heat sufficiently to generate these melts. In this model, the Schirmacher nappe in Dronning Maud Land could represent an extension of the aforementioned nappe to the south (e.g. Ravikant et al. 2004).
However, there are a number of possible problems with this model. First, the klippen overlying the Nampula and Dronning Maud Land crust are never seen to be intruded by the late-tectonic granites as might be expected. Second, the granitoids are typically charnockitic A-types and not the minimum-melt granitoids that might be expect to result from slowly heated crust. Finally, new dates indicate that there is significant young shearing along the Lurio Belt around 530–500 Ma (Bingen et al. 2006a, b), suggesting that the main movements on the structure are considerably younger than 620–550 Ma. (3) A third model, by the Norconsult Consortium (2007) infers an opposite, NW-directed tectonic transport direction, including thrusting and exhumation of the granulite-facies rocks at c. 620 – 550 Ma to the north of the Lurio Belt. This event was followed by a phase of extension at c. 530 – 490 Ma during which the Mesoproterozoic to Neoproterozoic basement terranes north of the Lurio Belt were juxtaposed with the Nampula Province along the Lurio Belt by extensional shearing, as indicated by the young metamorphic ages from the belt. Our present study sheds little light on the early exhumation history of the granulite-facies rocks but does contribute to the understanding of the later, mainly extensional history of the orogen. The late-tectonic granitoids intrude neither the granulite-facies klippen that overlie the Nampula Province nor the disconformably overlying latest Neoproterozoic metasedimentary rocks of the Alto Benfica and Mecuburi Groups. This suggests that the emplacement of the granulitefacies klippen and the Alto Benfica and Mecuburi Group metasediments occurred very late in the tectonic history, and is probably related to late orogenic collapse, which possibly continued up to c. 500 Ma. Furthermore, the late-tectonic granitoids represent high-T melts (charnockites, mangerites), which, at least in Dronning Maud Land, were immediately preceded by gabbroic intrusions. These high-T melts clearly require a major, possibly external, heat source. At the same time, the intruded country rocks show a steep isothermal decompression path, indicating rapid uplift without adequate cooling (Jacobs et al. 2003b; Colombo & Talarico 2004). This process can possibly be better explained by generation of the magmas in an elevated thermal environment brought about by partial delamination of the orogenic root, followed by the influx of hot asthenosphere and accompanied by rapid mechanical thinning. In this model the Lurio Belt could represent an accommodation zone between two thermomechanically very different parts of the orogen, rather than a suture zone.
GRANITES IN EAAO
Summary and conclusions A widespread late- to post-tectonic Cambrian magmatic province is recognized in the southern part of the EAAO, within northeastern Mozambique and central Dronning Maud Land, two areas that were thought to be contiguous within Gondwana. It covers an area of more than 15 000 km2, and would have stretched from the northern margin of the Nampula Complex (the Lurio Belt), in northeastern Mozambique, to central Dronning Maud Land, decreasing gradually westwards in volume to the eastern Sverdrupfjella, where the magmatism stops close to the frontal zone of the orogen. Extensional tectonics and late-tectonic magmatism is recorded between c. 530 and 485 Ma. The early stage (c. 530–520 Ma) is characterized by minor gabbro intrusions, followed by the main charnockite– granitoid magmatic event, which was short-lived between c. 510 and 500 Ma. The Cambrian plutons are mostly hosted within juvenile Mesoproterozoic (1.2–1.05 Ga) high-grade gneisses. The geochemistry of the granitoids is broadly ‘anorogenic’, ‘within-plate’ and ‘A2-type’ as defined from various geochemical discriminants. The intrusions are interpreted to have crystallized at mid-crustal levels during collapse and extension of the orogen, possibly accompanied by delamination of the lithosphere root. Hot asthenosphere, rising to the base of the lower crust above the subsiding orogenic root, would have provided the heat source for the magmatism, which is typically anhydrous, hightemperature charnockitic (especially in Dronning Maud Land). The large volumes of late-tectonic magmatism are confined to the southern third of the EAAO, bounded to the north by the Lurio Belt. The spatial extent of the magmatism is structurally controlled, which makes it unlikely that a simple mantle plume model could be invoked. The timing of the proposed collapse –extension event is constrained by new age data from metamorphic rims, which grew during shearing in synextensional late ductile mylonites. These rims, which surround cores with Mesoproterozoic protolith ages, were dated at 507 + 9 Ma, statistically indistinguishable from the ages of the main granitoid intrusions in Dronning Maud Land and northern Mozambique, which gave ages ranging from 512 to 499 Ma. This confirms that some of the extensional shearing is contemporaneous with the voluminous late-tectonic magmatism. Thus far, it is unclear whether the southdirected escape tectonic model of Jacobs & Thomas (2004) is related to the proposed delamination, as there are at present no robust age constraints for the escape tectonic event. The Lurio Belt is certainly an important boundary zone and might represent a mid- to lower-crustal
87
thermomechanical accommodation zone separating the southern part of the EAAO with a delaminated orogen root from crust to the north, where delamination did not occur. We acknowledge constructive reviews from G. Grantham, M. Owada and R. J. Stern. Part of the research was funded through a Heisenberg Fellowship to J.J., DFG Ja 617/16. J.J. also acknowledges support through a Gledden Fellowship at the Tectonics Special Research Centre (TSRC), University of Western Australia, Perth. The work in Mozambique is part of a Mineral Resources Management Capacity Building Project, Republic of Mozambique, financed by the Nordic Development Fund and the World Bank. U – Pb analyses were conducted using the SHRIMP II ion microprobe at the John de Laeter Centre of Mass Spectrometry in Perth, Australia, which is operated by a university – government consortium, with the support of the Australian Research Council, and also at the NORDSIM laboratory, operated and financed under an agreement between the research councils of Denmark, Norway and Sweden, the Geological Survey of Finland, and the Swedish Museum of Natural History. L. Ilynsky, K. Linde´n and M. Whitehouse are thanked for operating the NORDSIM laboratory. R.J.T. acknowledges permission to publish from the Executive Director, BGS. M.T.D.W. publishes with permission of the Executive Director, Geological Survey of Western Australia. This is TSRC Publication 414 and NORDSIM Publication 196.
References B AUER , W. & J ACOBS , J. 2005. Comment on ‘A geochronological and geochemical study of rocks from Gjelsvikfjella, Dronnng Maud Land, Antarctica— implications for Mesoproterozoic correlations and assembly of Gondwana’. Precambrian Research, 138, 177–179. B AUER , W., J ACOBS , J., F ANNING , C. M. & S CHMIDT , R. 2003. Late Mesoproterozoic arc and back-arc volcanism in the Heimefrontfjella (East Antarctica) and implications for the palaeogeography at the southeastern margin of the Kaapvaal– Grunehogna Craton. Gondwana Research, 6, 449– 465. B AUER , W., J ACOBS , J. & P AECH , H.-J. 2004. Structural evolution of the crystalline basement of central Dronning Maud Land, East Antarctica. Geologisches Jahrbuch, Reihe B, 96, 325– 363. B INGEN , B., V IOLA , G., H ENDERSON , I. H. C. ET AL . 2006a. Geochronology of Pan-African terrain assembly in the NE Mozambique. In: 21st Colloquium of African Geology, Maputo, Mozambique. Abstracts, 12–14. B INGEN , B., B OYD , R., T HOMAS , R. J. ET AL . 2006b. Crustal architecture in Northern Mozambique: results from a regional bedrock mapping project. In: EGU General Assembly, Vienna, Geophysical Research Abstracts, Vol. 8, 06049. B ISNATH , A., F RIMMEL , H. E., A RMSTRONG , R. A. & B OARD , W. S. 2006. Tectonothermal evolution of the Maud Belt: New SHRIMP U– Pb zircon data
88
J. JACOBS ET AL.
from Gjelsvikfjella, Dronning Maud Land, East Antarctica. Precambrian Research, 150, 95– 121. B LACK , L. P., K AMO , S. L., A LLEN , C. M., A LEINIKOFF , J. N., D AVIS , D. W., K ORSCH , R. J. & F OUDOULIS , C. 2003. TEMORA 1: a new zircon standard for Phanerozoic U –Pb geochronology. Chemical Geology, 200, 155– 170. B ONIN , B. 2007. A-type granites and related rocks: Evolution of a concept, problems and prospects. Lithos, 97, 1 –29. B UCHER , K. & F ROST , B. R. 1995. Charnockites and granulites of the Thor Range, Queen Maud Land, Antarctica—water recycling in high-grade metamorphism. Terra Abstracts, 7, 314. C OLOMBO , F. & T ALARICO , F. 2004. Regional metamorphism in the high-grade basement of central Dronning Maud Land, East Antarctica. Geologisches Jahrbuch, Reihe B, 96, 7 –47. E BY , G. N. 1992. Chemical subdivision of the A-type granitoids: petrogenetic and tectonic implications. Geology, 20, 641–644. E NGVIK , A. K. & E LVEVOLD , S. 2004. Pan-African extension and near-isothermal exhumation of a granulite facies terrain, Dronning Maud Land, Antarctica. Geological Magazine, 141, 649– 660. E NGVIK , A. K., B ERTRAM , A., K ALTHOFF , J. F., S TO¨ CKHERT , B., A USTRHEIM , H. & E LVEVOLD , S. 2005. Magma-driven hydraulic fracturing and infiltration of fluids into the damaged host rock, an example from Dronning Maud Land, Antarctica. Journal of Structural Geology, 27, 839– 854. E NGVIK , A. K., T VETEN , T., B INGEN , B., V IOLA , G., E RAMBERT , M., F EITO , P. & DE A ZAVEDO , S. 2007. P– T –t evolution and textural evidence for decompression of Pan-African high-pressure granulites, Lurio Belt, northeastern Mozambique. Journal of Metamorphic Geology, 25, 935– 952. F ROST , B. R. & B UCHER , K. 1993. Charnockites and granulites of the Thor Range, Queen Maud Land, Antarctica; the granulite uncertainty principle exemplified. Abstracts, Annual Meeting of the Geological Society of America, A-448. G RANTHAM , G. H., M ABOKO , M. & E GLINGTON , B. M. 2003. A review of the evolution of the Mozambique Belt and implications for the amalgamation and dispersal of Rodinia and Gondwana. In: Y OSHIDA , M., W INDLEY , B. F. & D ASGUPTA , S. (eds) Proterozoic East Gondwana: Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 401– 426. G RANTHAM , G. H., M ACEY , P. H., I NGRAM , B. A. ET AL . 2007. Terrane correlation between Antarctica, Mozambique and Sri Lanka; comparisons of geochronology, lithology, structure and metamorphism. In: C OOPER , A. K., R AYMOND , C. R. & ISAES Editorial Team (eds) Antarctica: A Keystone in the Changing World—Online Proceedings of the 10th International Symposium on Antarctic Earth Sciences. US Geological Society Open-File Report, 2007-1047, Extended Abstract 004. World Wide Web Address: http:// pubs.usgs.gov/of/2007/1047/. H ENJES -K UNST , F. 2004. Further evidence for PanAfrican magmatism and metamorphism in Central Dronning Maud Land, East Antarctica, from rocks
at Schirmacheroase: a geochronological study. Geologisches Jahrbuch, Reihe B, 96, 255– 292. J ACOBS , J. & T HOMAS , R. J. 2002. The Mozambique Belt from an East Antarctic perspective. In: G AMBLE , J. A., S KINNER , D. N. B. & H ENRYS , S. (eds) Antarctica at the Close of a Millennium. Proceedings of the 8th International Symposium on Antarctic Earth Sciences. Royal Society of New Zealand Bulletin, 35, 3 –18. J ACOBS , J. & T HOMAS , R. J. 2004. A Himalayan-type indenter-escape tectonic model for the southern part of the Late Neoproterozoic/Early Paleozoic East African–Antarctic Orogen. Geology, 32, 721– 724. J ACOBS , J., F ANNING , C. M., H ENJES -K UNST , F., O LESCH , M. & P AECH , H.-J. 1998. Continuation of the Mozambique Belt into East Antarctica. Grenvilleage metamorphism and polyphase Pan-African highgrade events in central Dronning Maud Land. Journal of Geology, 106, 385– 406. J ACOBS , J., B AUER , W. & F ANNING , C. M. 2003a. Late Neoproterozoic/Early Palaeozoic events in central Dronning Maud Land (East Antarctica) and significance for the southern extension of the EAO into East Antarctica. Precambrian Research, 126, 27– 53. J ACOBS , J., K LEMD , R., F ANNING , C. M., B AUER , W. & C OLOMBO , F. 2003b. Extensional collapse of the Late Neoproterozoic/Early Palaeozoic East African/ Antarctic Orogen in central Dronning Maud Land. In: Y OSHIDA , M., W INDLEY , B. F. & D ASGUPTA , S. (eds) Proterozoic East Gondwana: Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 271–287. J ACOBS , J., P ISAREVSKY , S., T HOMAS , R. J. & B ECKER , T. 2008. The Kalahari Craton during the assembly and dispersal of Rodinia. Precambrian Research, 160, 142–158. J OSHI , A., P ANT , N. C. & P ARIMOO , M. L. 1991. Granites of the Petermann Ranges, East Antarctica, and implications on their genesis. Journal of the Geological Society of India, 36, 169– 181. K LIMOV , L. V., R AVICH , M. G. & S OLOVIEV , D. S. 1964. Charnockites of East Antarctica. In: A DIE , R. J. (ed.) Antarctic Geology. North Holland, Amsterdam, 455–462. K RO¨ NER , A., S ACCHI , R., J AECKEL , P. & C OSTA , M. 1997. Kibaran magmatism and Pan-African granulite metamorphism in northern Mozambique: single zircon ages and regional implications. Journal of African Earth Science, 25, 467–484. L I , Z., T AINOSHO , Y., K IMURA , J., S HIRAISHI , K. & O WADA , M. 2003. Pan-African alkali granitoids from the Sør Rondane mountains, East Antarctica. Gondwana Research, 6, 595–605. L OISELLE , M. C. & W ONES , D. R. 1979. Characteristics and origin of anorogenic granites. Abstracts, Geological Society of America Annual Meeting, 11, 468. L UDWIG , K. R. 1998. On the treatment of concordant uranium– lead ages. Geochimica et Cosmochimica Acta, 62, 665–676. L UDWIG , K. R. 2001. Squid 1.02, A User’s Manual. Berkeley Geochronology Center, Special Publications, 2. L UDWIG , K. R. 2003. User’s Manual for Isoplot 3.00: A Geochronological Toolkit for Microsoft Excel. Berkeley Geochronology Center, Special Publications, 4.
GRANITES IN EAAO M ARKL , G. & H ENJES -K UNST , F. 2004. Magmatic conditions of formation and autometasomatism of post-kinematic charnockites in central Dronning Maud Land, East Antarctica: a model of magmatic evolution. Geologisches Jahrbuch, Reihe B, 96, 139–185. M IKHALSKY , E. V., B ELIATSKY , B. V., S AVVA , E. V., W ETZEL , H.-U., F EDOROV , L. V., W EISER , T. & H AHNE , K. 1997. Reconnaissance geochronologic data on polymetamorphic and igneous rocks of the Humboldt Mountains, Central Queen Maud Land, East Antarctica. In: R ICCI , C. A. (ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antartica, Sienna, 45– 54. M OYES , A. B., G ROENEWALD , P. B. & B ROWN , R. W. 1993. Isotopic constraints on the age and origin of the Brattskarvet intrusive suite, Dronning Maud Land, Antarctica. Chemical Geology (Isotope Geosciences Section), 106, 453–466. NORCONSULT CONSORTIUM 2007. Mineral resources management capacity building project, Republic of Mozambique, Component 2: Geological infrastructure development project, Geological Mapping Lot 1, Report B6.f. O HTA , Y., T ØRUDBAKKEN , B. O. & S HIRAISHI , K. 1990. Geology of Gjelsvikfjella and western Mu¨hligHofmannfjella, Dronning Maud Land, east Antarctica. Polar Research, 8, 99– 126. O WADA , M., B ABA , S., L A¨ UFER , A. L., E LVEVOLD , S., S HIRAISHI , K. & J ACOBS , J. 2003. Geology of eastern Mu¨hlig-Hofmannfjella and Filchnerfjella in Dronning Maud Land, East Antarctica: A preliminary report on a Japan–Norway– Germany joint geological investigation. Polar Geoscience, 16, 108–136. P AECH , H.-J. 2001. Pervasive Pan-African reactivation of the Grenvillian crust and large igneous intrusions in central Dronning Maud Land, East Antarctica. In: M ILLER , J. A., H OLDSWORTH , R. E., B UICK , I. S. & H AND , M. (eds) Continental Reactivation and Reworking. Geological Society, London, Special Publications, 184, 343–355. P AECH , H.-J., J ACOBS , J., B AUER , W. ET AL . 2004. Geological map 1:500 000 of the central Dronning Maud Land; Explanatory notes to the 1:500 000 geological map of central Dronning Maud Land. Geologisches Jahrbuch, Reihe B, 96, 97. P AULSSON , O. & A USTRHEIM , H. 2003. A geochronological and geochemical study of rocks from Gjelsvikfjella, Dronnng Maud Land, Antarctica— implications for Mesoproterozoic correlations and assembly of Gondwana. Precambrian Research, 125, 113–138. P INNA , P., J OURDE , G., C ALUEZ , J.-Y., M ROZ , J. P. & M ARQUES , J. M. 1993. The Mozambique Belt in northern Mozambique: Neoproterozoic (1100–850 Ma) crustal growth and tectogenesis and superimposed Pan-African (800–550 Ma) tectonism. Precambrian Research, 62, 1–59. R AVICH , M. G. & K AMENEV , E. N. 1975. Crystalline Basement of the Antarctic Platform. Wiley, New York. R AVICH , M. G. & S OLOVIEV , D. S. 1966. Geology and petrology of the central part of the mountains
89
of central Dronning Maud. Trudy NauchnoIsseldovatel’skogo Instituta Geologii Arktiki, 141 [in Russian]. R AVIKANT , V., R AO , Y. J. B. & G OPALAN , K. 2004. Schirmacher Oasis as an extension of the Neoproterozoic East African Orogen into Antarctica: new Sm–Nd isochron age constraints. Journal of Geology, 112, 607–616. R OLAND , N. W. 2002. Pan-African granitoids in central Dronning Maud Land, East Antarctica. Petrography, geochemistry and plate tectonic setting. In: G AMBLE , J. A., S KINNER , D. N. B. & H ENRYS , S. (eds) Antarctica at the Close of a Millenium. Royal Society of New Zealand Bulletin, 35, 85– 91. R OLAND , N. W. 2004a. Pan-African granites in central Dronning Maud Land, East Antarctica: Product of collision or intra-plate event? Zeitschrift der Deutschen Geologischen Gesellschaft, 154, 469– 479. R OLAND , N. W. 2004b. Pan-African granite– charnockite magmatism in central Dronning Maud Land, East Antarctica: petrography, geochemistry and plate tectonic implications. Geologisches Jahrbuch, Reihe B, 96, 187– 231. S HIRAISHI , K., A SAMI , M. & O HTA , Y. 1983. Geology and petrology of the Yamato Mountains. In: O LIVER , R. L., J AMES , P. R. & J AGO , J. B. (eds) Antarctic Earth Science. Cambridge University Press, Cambridge, 50–53. S HIRAISHI , K., E LLIS , D. J., H IROI , Y., F ANNING , C. M., M OTOYOSHI , Y. & N AKAI , Y. 1994. Cambrian orogenic belt in east Antarctica and Sri Lanka. Implications for Gondwana assembly. Journal of Geology, 102, 47– 65. S TACEY , J. S. & K RAMERS , J. D. 1975. Approximation of terrestrial lead isotope evolution by a two-stage model. Earth and Planetary Science Letters, 26, 207– 221. S TERN , R. J. 1994. Arc assembly and continental collision in the Neoproterozoic East African Orogen: implications for the consolidation of Gondwana. Annual Reviews of Earth and Planetary Sciences, 22, 319–351. T HOMAS , R. J., B AUER , W., B INGEN , B. ET AL . 2006. Mozambique belt in the Milange–Mocuba–Malema area, Moc¸ambique. 21st Colloquium of African Geology, Maputo, Mozambique, Abstracts, 161. V AN A UTENBOER , T. & L OY , W. 1972. Recent geological investigations in the Sør Rondane mountains, Belgicafjella and Sverdrupfjella, Dronning Maud Land. In: A DIE , R. J. (ed.) Antarctic Geology and Geophysics. Universitetsforlaget, Oslo, 563–571. V IOLA , G., H ENDERSON , I., B INGEN , B., F EITIO , P., T HOMAS , R., H OLLICK , L. & J ACOBS , J. 2006. A new tectonic framework for northern Mozambique. 21st Colloquium of African Geology, Maputo, Mozambique. Abstracts, 168 –170. W HITEHOUSE , M. J. & K AMBER , B. S. 2005. Assigning dates to thin gneissic veins in high-grade metamorphic terranes: a cautionary tale from Akilia, Southwest Greenland. Journal of Petrology, 46, 291– 318. W HITEHOUSE , M. J., K AMBER , B. S. & M OORBATH , S. 1999. Age significance of U–Th– Pb zircon data from early Archaean rocks of west Greenland—a
90
J. JACOBS ET AL.
reassessment based on combined ion-microprobe and imaging studies. Chemical Geology, 160, 201– 224. W IEDENBECK , M., A LLE´ , P., C ORFU , F. ET AL . 1995. Three natural zircon standards for U–Th–Pb, Lu–Hf, trace element and REE analyses. Geostandards Newsletter, 19, 1–23.
W ILLIAMS , I. S. 1998. U– Th–Pb geochronology by ion microprobe. In: M C K IBBEN , M. A., S HANKS , W. C., III & R IDLEY , W. I. (eds) Applications of Microanalytical Techniques to Understanding Mineralisation Processes. Reviews of Economic Geology, 7, 1 –35.
Terrane correlation between Antarctica, Mozambique and Sri Lanka; comparisons of geochronology, lithology, structure and metamorphism and possible implications for the geology of southern Africa and Antarctica G. H. GRANTHAM1, P. H. MACEY2, B. A. INGRAM1, M. P. ROBERTS3,4, R. A. ARMSTRONG5, T. HOKADA6, K. SHIRAISHI6, C. JACKSON7, A. BISNATH8 & V. MANHICA9 1
Council for Geoscience, P/Bag X112, Pretoria, South Africa (e-mail:
[email protected]) 2
Council for Geoscience, Bellville, South Africa
3
MSSP-Geomap Project, c/o Geological Survey of Papua New Guinea, Port Moresby, Papua New Guinea 4
Council for Geoscience, Walmer, Port Elizabeth, South Africa
5
RSES, Australian National University, Canberra, A.C.T. 0200, Australia 6
National Institute of Polar Research, Itabashi, Tokyo, Japan
7
51 Saint David’s Road, Claremont, Cape Town, South Africa
8
Caracle Creek International Consulting Inc., Johannesburg, South Africa 9
Direca˜o Nacional Geologia, Maputo, Mozambique
Abstract: Analysis of new lithological, structural, metamorphic and geochronological data from extensive mapping in Mozambique permits recognition of two distinct crustal blocks separated by the Lurio Belt shear zone. Extrapolation of the Mozambique data to adjacent areas in Sri Lanka and Dronning Maud Land, Antarctica permits the recognition of similar crustal blocks and allows the interpretation that the various blocks in Mozambique, Sri Lanka and Antarctica were once part of a mega-nappe, forming part of northern Gondwana, which was thrust-faulted c. 600 km over southern Gondwana during amalgamation of Gondwana at c. 590–550 Ma. The data suggest a deeper level of erosion in southern Africa compared with Antarctica. It is possible that this thrust domain extends, through the Zambezi Belt or Valley, as far west as the Damara Orogen in Namibia with the Naukluft nappes in Namibia, the Makuti Group, the Masoso Suite in the Rushinga area and the Urungwe klippen in northern Zimbabwe, fitting the mega-nappe pattern. Erosional products of the mountain belt are now represented by 700– 400 Ma age detrital zircons present in the various sandstone formations of the Transantarctic Mountains, their correlatives in Australia, as well as the Urfjell Group (western Dronning Maud Land) and probably the Natal Group in South Africa.
‘What’s in a name? That which we call a rose by any other name would smell as sweet’ (Romeo and Juliet (II, ii, 1–2), Shakespeare, c. 1595). This paper could equally have been titled ‘The nature and extent of the Lurio Belt inferred from the geochronology, structure, lithology and metamorphic histories of adjacent crustal blocks’ or alternatively ‘The errant hitchhiker terranes of northern Gondwana’.
In 2000 an ambitious project to map Mozambique was initiated by the Mozambique Government The project was funded by the World Bank, the Nordic Development Fund and the governments
of South Africa and Mozambique. The Norwegian Geological Survey (NGU) and British Geological Survey (BGS) consortium assumed responsibility for mapping most of northeastern Mozambique (Fig. 1). A consortium including the Finnish Geological Survey (GTK) and a private company assumed responsibility for most of northwestern, central and southern Mozambique (Fig. 1) whereas the Council for Geoscience of South Africa assumed responsibility for the mapping of eleven 11 degree sheets (Fig. 1). Nine of these
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 91– 119. DOI: 10.1144/SP308.4 0305-8719/08/$15.00 # The Geological Society of London 2008.
92
G. H. GRANTHAM ET AL.
Fig. 1. Map of Mozambique showing the areas mapped by the various groups. CGS, Council for Geoscience of South Africa; NGU/BGS, Norwegian Geological Survey– British Geological Survey; GTK, Finnish Geological Survey.
sheets were located in NE Mozambique and two in NW Mozambique. The funding for mapping of the areas of northern Mozambique mapped by the NGU–BGS and GTK teams came from the Nordic Development Fund. The project is now complete and the potential implications of some of the new data gathered in the mapping exercise are presented in summarized form and interpreted here. A review of the basement geology of Mozambique (Grantham et al. 2003) highlighted the paucity of reliable geochronological data in Mozambique itself (approximately five single zircon determinations at the time), in contrast to neighbouring areas. It also highlighted that a potential crustal boundary existed between the apparently juvenile southern end of the Mozambique Belt and an older block in northern Mozambique, with the northern end having limited juvenile rocks with extensive reworking of Palaeoproterozoic and Archaean rocks. The current project along with other studies has resulted in the number of new, reliable zircon ages now exceeding 100. Some of these data are summarized here and interpreted along with data from surrounding areas in a Gondwana context. These data have been critical for the definition of the model presented here. Recognizing the reconnaissance nature of the mapping, the application of low- to high-resolution magnetic and
radiometric aerial surveys and Landsat data has also contributed significantly to the new interpretation presented here by confirming the Lurio Belt as a prominent geophysical and geological structure. The Lurio Belt is manifested in the field as a north to NW, intermediate to steeply dipping, zone of high strain. The more exact nature of the Lurio Belt (i.e. thrust zone, strike-slip zone, zone of pure shear?) is the subject of current debate and has not been completely resolved by the mapping programme in Mozambique. The Lurio Belt (LB) was first described by Jourde & Vialette (1980), who described it as a suture of a major Lurian orogen with diverging nappes to the north and south (Pinna et al. 1993). Later interpretations by Pinna and others (Pinna 1995) considered it to be a late southerly thrust synform. Pinna et al. (1993) and Sacchi et al. (2000) proposed extensions of the Lurio structure into the Zambezi Belt, both of which involved late south-directed reverse thrusting. Jamal (2005) described a complex history for the Lurio Belt, reporting that it has been affected by four phases of deformation under granulite- to amphibolite-facies conditions. The history recorded by Jamal (2005) is summarized as follows. An earlier deformation (D1) along the belt is represented by felsic segregations that commonly trace the D2/F2 folds. Map- and outcrop-scale F2 NE– SW-oriented isoclinal folds generally have subhorizontal fold axes. Fold attitudes are described as varying from subvertical upright to NW-dipping. Associated fold asymmetry suggests that these structures represent SE-directed thrusting. Continuous NW– SE shortening, with a dextral shear component, led to the refolding of the D2 isoclines about the F3 open to tight subvertical folds and the formation of an S3 axial-plane cleavage. D3 is interpreted as involving strike-slip shear under amphibolite-facies P–T conditions. S– C0 mylonitic fabrics observed throughout the Lurio Belt suggest a transpressive regime in response to an oblique compression. Thus the NE– SW Lurio Belt probably acted as a shear zone that accommodated high strain during a collision event that affected NE Mozambique. Viola et al. (2006) have recently questioned the interpretation of the Lurio Belt as a major suture. They reported that structures along the belt vary greatly, involving intense linear structures in the NE, becoming wider and less belt-like in the SW. They described tight to isoclinal folds with NNW-dipping axial planes and roughly down-dip stretching lineations. No clear kinematic indicators were observed. Strain accommodation, involving folding and conjugate shear zones within the Lurio Belt, is more intense than in the surrounding rocks. Evidence of SSE–NNW-directed regional
ANTARCTIC–MOZAMBIQUE –SRI LANKA CORRELATION
compression is pervasive. Viola et al. concluded that the Lurio Belt represents a belt of repeated activity and reworking and that the last strain increment reflects pure shear bulk flattening of the belt, lacking significant regional belt parallel simple shear. In contrast to the south-directed transport direction, they have inferred extensional collapse toward the WNW. Various workers have proposed that the Lurio Belt extends into Sri Lanka (Kro¨ner 1991, 2001; Grantham et al. 2003) and is represented there by the shear zone separating the Highland Complex from the Vijayan Complex. This correlation is supported by the proximity of Sri Lanka to northern Mozambique suggested by various Gondwana reconstructions (Lawver et al. 1998; Reeves & de Wit 2000). This paper summarizes lithological, geochronological, structural and metamorphic data and interprets them to suggest that the Lurio Belt represents a deep-crustal terrane boundary in northern Mozambique and that its possible extensions into Sri Lanka and Dronning Maud Land, Antarctica, as a low-angle thrust nappe complex, permit the recognition of various crustal blocks or terranes separated by correlatives of the Lurio Belt in varying attitudes.
Crustal structure of Mozambique Interpretation of aeroradiometric and aeromagnetic data supported by reconnaissance ground mapping by the various mapping teams has facilitated the recognition that NE Mozambique is divided into two dominant blocks separated by the Lurio Belt (Fig. 2) and its possible extensions westwards. For the purposes of this paper the block north of the Lurio Belt is termed the Namuno Block and that south of the Lurio Belt the Nampula Block. The Namuno Block comprises an accretionary stack of thrust-faulted complexes (Fig. 1) of varying age (Bingen et al. 2006; Bjerkgard et al. 2006; Viola et al. 2006). Mesoproterozoic complexes include the Unango, Marrupo, Naroto, Meluco and Angonia Complexes whereas Neoproterozoic complexes include the Xixano, Montepuez, Lalamo, M’Sawize and Muaquia Complexes (Bingen et al. 2006; Hollick et al. 2006; Thomas et al. 2006; Viola et al. 2006; Grantham et al. 2007a; Fig. 2). The thrust-faulted accretionary stack defined by these workers can be extended further west, via Malawi, to northwestern Mozambique, with Mesoproterozoic rocks of the Southern Irumide Belt forming the footwall to similar-age high-grade gneisses thrust westwards (Grantham et al. 2007a). The Southern Irumide Belt is similarly interpreted to be underlain by late Palaeoproterozoic basement
93
intruded by continental margin arc-related magmas between 1.09 and 1.4 Ga and strongly overprinted during the Pan-African Orogen (Johnson et al. 2006). The Southern Irumide Belt is also interpreted to comprise four shear zone bounded terranes, with the bounding faults having NW–SE-oriented strikes (Johnson et al. 2006). The Namuno Block thrust stacks recognized by Bingen et al. (2006) and Viola et al. (2006) and the Angonia Complex (Grantham et al. 2007a) are reported to have involved top-to-the-west and -WNW deformation whereas those further west are inferred to involve top-to-the-SW orientation (Johnson et al. 2006). It is readily apparent from the geophysical data that the whole thrust stack comprising the Mesoproterozoic to Neoproterozoic rocks from the Mozambique coast to the Angonia Complex in the west is itself sheared and rotated, with an apparent sinistral sense of rotation, where these rocks merge with the Lurio Belt and its possible extensions in the south. This indicates that the intense ductile strain deformation recorded in the north- to NW-dipping Lurio Belt either post-dated the amalgamation of the various complexes in the Namuno Block or was part of a synchronous, larger-scale, sinistral transpressional structure, with the ENE – WSW-oriented Lurio Belt being the central main shear zone bounded on the north by a westward (sinistral) directed accretionary stack. In contrast, the radiometric and aeromagnetic data for the area south of the Lurio Belt, the Nampula Block, do not show the same accretionary stack configuration. The Nampula Block is dominated by medium-grade migmatitic tonalitic orthogneisses and paragneisses and quartzofeldspathic orthogneisses that are complexly interfolded and intruded by undeformed to locally weakly deformed granitic intrusions. At least two granulite-grade klippen, the Monapo Complex and the Mugeba Complex (Pinna et al. 1993) are recognized overlying the Nampula Block. These two klippen contain high-grade granulite ortho- and paragneisses and have been regarded as remnants of a larger thrust sheet or sheets (Pinna et al. 1993). The granulite-grade klippen suggested by Pinna et al. (1993) in the vicinity of Nampula is not a klippen, but rather an area of sporadic in situ charnockitization that probably developed through the action of late-tectonic fluids (Macey et al. 2007). The Nampula Block is also partially transgressed by the north–south-oriented mylonitic Namama sinistral strike-slip shear zone (Cadoppi et al. 1987) in the SW (Fig. 2). The Namama Shear Zone appears to curve and disappear into NE –SW-oriented layer-parallel structures at its northern end. The lack of continuity of the Namuno Block nappe complex across the Lurio Belt implies that
94
G. H. GRANTHAM ET AL.
Fig. 2. Map showing the crustal structure of northern Mozambique and adjacent areas. The location of the sinistral Namama Shear Zone (Nm) is shown east of Mocuba in the south of the Nampula Block and the approximate location of the Geci Group (Gc) is shown in far northern Mozambique east of Malawi. The map is compiled from Barr & Brown (1987), Bingen et al. (2006), Bjerkgard et al. (2006), Hollick et al. (2006), Thomas et al. (2006), Grantham et al. (2007a, b) and Macey et al. (2007).
the Lurio Belt is not just a crustal shear zone that has segmented a uniform block of crust but represents a fundamental boundary along which two different blocks have been juxtaposed. The exact location of extensions of the Lurio structure westwards through southern Malawi into NW Mozambique and on into the Zambezi Valley are less clear, as a result of much of the area being overlain by Karoo-age sedimentary and volcanic rocks as well as being transected by the southern extensions of the western limb of the East
African Rift. Pinna et al. (1993) Pinna (1995), Sacchi et al. (2000) and Grantham et al. (2003) have supported an extension of the Lurio Belt into the Zambezi Belt. During the current mapping programme it was recognized by Koistinen et al. (2006) and Westerhof (2006) that the Tete Complex is allochthonous and forms the hanging wall of a large southward-directed thrust fault. These workers also record a large southward-directed c. east–west shear zone c. 10 km south of Tete. Barr & Brown
ANTARCTIC–MOZAMBIQUE –SRI LANKA CORRELATION
(1987) also reported a major east–west-oriented shear zone, the Sanangoe thrust zone c. 40 km north of Tete with a top-to-the-north direction of transport. It is possible that these structures represent extensions of the Lurio Belt westwards. In conclusion, the mapping programme has facilitated the definition of the Lurio Belt as a major crustal boundary, which will be used to interpret the geochronological, structural and lithological variations described below.
Rock types Namuno Block The composition of the Unango Complex varies widely and consists of granitic gneisses, some locally charnockitic, with biotite–hornblende gneisses and quartzite. The metamorphic grade varies from amphibolite to granulite grade and the rocks are extensively migmatized (Bjerkgard et al. 2006). Aeroradiometric and aeromagnetic surveys suggest that the rocks of the Unango Complex continue southwestwards through Malawi and become the supracrustal Angonia Complex in NW Mozambique west of the Malawi border. In this area interlayered, dominantly quartzofeldspathic gneisses with subordinate tonalitic and metabasic gneisses have been reported (Grantham et al. 2007a). Rare metapelitic gneisses and marbles are also seen. The metabasic gneisses have chemistry typical of enriched mid-ocean ridge basalt (E-MORB) rocks (Grantham et al. 2007a). Intruded into the supracrustal rocks are monzonites, syenites, anorthosite and pyroxenites of mostly uncertain age. The supracrustal sequences have ages of c. 1100–1050 Ma whereas an undeformed to weakly deformed monzonite has been dated at c. 560 Ma (Grantham et al. 2007a). Metamorphic overprints dated at c. 550 Ma have been reported from zircon rims and metamorphic titanite (Grantham et al. 2007a). The Angonia Group gneisses are thrust-faulted over granites of the Southern Irumide Complex to the east (Grantham et al. 2007a). The Marrupa Complex is dominated by granitic to tonalitic gneiss, with mafic amphibolitic orthogneisses, quartzite and quartz–feldspar gneiss (Bjerkgard et al. 2006). The rocks are characterized by amphibolite-facies mineralogy. The geochemistry of the orthogneisses suggests that they are mediumto high-K calc-alkaline rocks with SiO2 ranging between c. 42 and 78 wt% and K2O ranging between 0.3 and 6.1 wt% (Bjerkgard et al. 2006). The Nairoto Complex consists of variably migmatized granitic to tonalitic orthogneisses with calc-alkaline compositions (Bjerkgard et al. 2006). Mineral assemblages
95
are typical of amphibolite-facies metamorphism (Bjerkgard et al. 2006). The Meluco Complex comprises mostly granitic to granodioritic orthogneisses. Mineral assemblages are typical of amphibolite-facies metamorphism (Bjerkgard et al. 2006). The Xixano Complex includes part of the Chiure Supergroup and autochthonous supracrustal gneisses described by Pinna et al. (1993), and comprises a variety of paragneisses including marble, biotite gneiss, mica schists, meta-arenites, granitic to tonalitic gneisses and amphibolites (Bjerkgard et al. 2006). The metamorphic grade within the Xixano Complex is amphibolite facies to granulite facies (Bjerkgard et al. 2006). The Muaquia Complex comprises granitic, tonalitic and gabbroic gneisses, amphibolites, mica gneiss and calc-silicate gneisses, and is predominantly mafic to intermediate in composition (Bjerkgard et al. 2006). The rocks have mineralogy typical of dominantly amphibolite-facies metamorphism (Bjerkgard et al. 2006). The M’Sawize Complex comprises banded migmatitic gneisses, granulitic gneiss and mafic granulite with mineralogy typical of granulitefacies metamorphism (Bjerkgard et al. 2006). The M’Sawize Complex comprises part of the Msawize Group of Pinna et al. (1993), who included their unit as part of the Lurio Supergroup. The Lalamo Complex contains a variety of paragneisses including marble, biotite gneiss, mica schists, meta-arenites and granitoid gneisses with mineralogy typical of amphibolite-facies metamorphism (Bjerkgard et al. 2006). The Montepuez Complex was previously defined as part of the large Chiure Group by Pinna et al. (1993) and contains granitic to granodioritic gneiss, biotite gneiss and marbles with mineral parageneses typical of amphibolite-facies metamorphism (Bjerkgard et al. 2006). The Ocua Complex comprises rocks defined as the Lurio Supergroup by Pinna et al. (1993) and consists of a tectonic me´lange (Thomas et al. 2006). The main rock types are mostly granulitic gneisses of tonalitic, dioritic and granitic composition, amphibolitic and granulitic gneisses as well as ultramafic and metaluminous gneiss (Bjerkgard et al. 2006). The high strains characteristic of the eastern Lurio Belt become less distinct to the SW (Viola et al. 2006).
Nampula Block The description of the rock units of the Nampula Block is summarized from Macey et al. (2007), and Grantham et al. (2007b), who confirmed descriptions by earlier workers including Pinna et al. (1993) and Sacchi et al. (1984) amongst others. Six lithostratigraphic units are recognized in the Nampula Block. They comprise the Mesoproterozoic gneisses of the Mocuba Suite, the Culicui
96
G. H. GRANTHAM ET AL.
Suite, the Rapale Gneiss, the Mamala Gneiss, the Molocue Group and the Cambrian granites of the Murrupula Suite. The Mocuba Suite consists dominantly of migmatitic banded tonalitic gneisses and subordinate mafic rocks with amphibolitefacies mineralogy. Compositions are dominantly calc-alkaline. Crystallization ages of c. 1125 Ma have been recorded (Macey et al. 2007). Not only does the strongly migmatized character of the Mocuba Suite distinguish it from the other Mesoproterozoic gneisses, it also indicates that a Mesoproterozoic orogenic event was experienced by these rocks of the Nampula Block. The Rapale Gneiss is of similar tonalitic composition but is clearly intrusive into the Mocuba Suite and has crystallization ages of c. 1090 Ma. The Culicui Suite is dominated by megacrystic, typically highly sheared, augen gneisses, which locally, in low-strain zones, preserve charnockitic mineralogy. In general, however, the metamorphic assemblages are typical of amphibolite-facies metamorphism. Crystallization ages from the Culicui Suite range typically between c. 1070 and 1090 Ma. The Mamala Gneiss is relatively uniform equigranular medium- to fine-grained leuco-quartzofeldspathic gneiss with uniform field and geophysical characteristics. Crystallization ages from the Mamala Gneiss are c. 1090 Ma. The Molocue Group comprises a banded interlayered sequence of dominantly quartzo-feldspathic para- and orthogneisses with subordinate amphibolites and calc-silicates. Three younger pre-Gondwana breakup lithological entities are recognized in the Nampula Block: the Murrupula Suite, the Mugeba Complex and the Monapo Complex. The Mugeba Complex is dominated by granulite-grade garnet–pyroxene intermediate orthogneisses with subordinate garnet– pyroxene metabasic granulites and garnet–sillimanite–rutile metapelitic gneiss. The Monapo Complex contains mostly granulite-grade banded supracrustal gneisses with subordinate Grt–Sil–Rtbearing metapelites. Intruded into the supracrustal gneisses are weakly deformed to apparently undeformed clinopyroxenites, syenites, granite and carbonatite (Siegfried 1999; Grantham et al. 2007b). Crystallization ages of c. 635 Ma are recognized from the Monapo Complex (Jamal 2005; Grantham et al. 2007b), whereas discordant zircons from the Mugeba Complex suggest a c. 1000 Ma protolith. Metamorphic ages from these complexes are c. 635 Ma and c. 580 Ma (Grantham et al. 2007b; Macey et al. 2007). The Murrupula Suite comprises granitoid intrusions, mostly undeformed and emplaced as kilometre-scale plutons to metre-scale pegmatitic dykes (Macey et al. 2007). The compositions vary from syenitic to granitic and include equigranular medium-grained varieties to coarse-grained
porphyritic types. The chemistry of the intrusions varies from metaluminous A-type rocks to peraluminous mica granites (Macey et al. 2007). The A-type intrusions have chemistries typical of A2 granites (Eby 1992), which are interpreted as typically being generated post-orogenically (Bonin 2007) and from continental crust that has been through a cycle of continent–continent collision or island arc magmatism (Eby 1992). Crystallization ages of the intrusions vary from c. 495 to c. 530 Ma.
Discussion On a purely lithological composition basis there is little to indicate a major crustal boundary defined by the Lurio Belt. Broadly, the Namuno Block and the Nampula Block are dominantly underlain by quartzofeldspathic gneisses, with the Lurio Belt and related Ocua Complex rocks being characterized by strong geophysical signatures and evidence of high strain in the field. The reduced geophysical signature of the Lurio Belt in the SW could possibly result from the structure attaining a shallower dip and being folded. This possibility requires additional investigation. Lithological distinguishing factors between the Namuno and Nampula Blocks include (1) a higher prevalence of supracrustal rocks containing metapelites and marbles north of the Lurio Belt as well as (2) the subordinate, but significant, presence of relatively alkaline syenitic orthogneisses north of the Lurio Belt. The only lithological exceptions to these distinguishing factors are found in the Monapo and Mugeba Complex klippen overlying the Nampula Block.
Geochronology of Mozambique and surrounding areas In comparing the geochronology of the different areas, histograms with bin sizes of 50 Ma are used in conjunction with probability density distributions. The limitations of these two methods have been described in detail by Sircombe (2000). The limitation of the use of histograms alone is that they do not take the error estimate of an age into account whereas a limitation in the use of probability density distributions alone is that they do not quantify the number of ages recorded within a particular age or bin range.
Southeastern Africa Figure 3 shows igneous crystallization and metamorphic ages from the Namuno Block, comprising the area north of the Lurio Belt in NE Mozambique
ANTARCTIC–MOZAMBIQUE –SRI LANKA CORRELATION
97
Fig. 3. (a, b) Histograms and probability density curves of crystallization ages from the Namuno Block (a), comprising an area north of the Lurio Block (NLB), a southern Irumide Belt Block (SIB) and Malawi (Mal), and (b) Nampula Block. (c, d) Histograms and probability density curves of metamorphic ages from the Namuno Block (c), comprising an area north of the Lurio Block (NLB), a southern Irumide Belt Block (SIB) and Malawi (Mal), and (d) Nampula Block igneous crystallization and metamorphic zircons. The lines for 600 Ma and 1100 Ma are shown for reference in all the figures.
(NLB), Malawi (Mal) and the southern Irumide areas of southern Zambia and NW Mozambique (SIB) and the Nampula Block (see Fig. 2). Data available include those generated during the mapping programme (Bingen et al. 2006; Ma¨ntta¨rri et al. 2006; Grantham et al. 2007a, b; Macey et al. 2007) along with additional data from Costa et al. (1994), Kro¨ner et al. (1997, 2001), Sacchi et al. (2000), Manhica et al. (2001), Jamal (2005) and Johnson et al. (2005, 2006). The data are dominantly derived from sensitive high-resolution ion microprobe (SHRIMP), inductively coupled plasma mass spectrometry (ICP-MS) or thermal ionization mass spectrometry (TIMS) analyses, except for the Pb– Pb evaporation data of Kroner et al. (2001) from southern Malawi. A few mineral – whole-rock Sm – Nd data are included from the Southern Irumide Block. In addition, only the data from the Southern Irumide Belt of Johnson et al. (2005, 2006) have been used for the geochronological comparisons below. The data used for the crystallization age
analysis from the Namuno Block are shown in Table 1 whereas the age data from the Nampula Block are summarized in Table 2. Comparison of the histograms and probability density distribution curves shows that all areas have broadly extensive Mesoproterozoic crystallization ages of c. 900–1150 Ma as well as Neoproterozoic–Cambrian ages of c. 650–450 Ma. However, a significant difference is that the Namuno Block (Fig. 3a) is characterized by crystallization ages from c. 650 to 900 Ma, which are absent from the Nampula Block (Fig. 3b). In the Nampula Block three ages of c. 630 Ma are recorded from the Monapo (two) and Mugeba klippen (Jamal 2005; Grantham et al. 2007b; Macey et al. 2007). These data clearly demonstrate that the Monapo and Mugeba klippen have age characteristics similar to rocks exposed north of the Lurio Belt in the Namuno Block. Other rock types with ages of c. 800 Ma possibly located south of Lurio Belt extensions are the granitic rocks intruded into the Rushinga area of NE Zimbabwe and western Mozambique
98
G. H. GRANTHAM ET AL.
Table 1. Crystallization age data used to construct figures for the Namuno Block from the southern Irumide Belt and Malawi Unit and sample number Biotite–hornblende gneiss (MA16) Pelitic paragneiss (MA8) Chewore Ophiolite plagiogranite (sample SJ106.1) Kaourera Arc meta-dacite (sample SJ220) Kadunguri Whiteschists Chewore Inlier Granulite Terrane (sample ADC) Chewore Inlier Zambezi Terrane orthogneiss (sample AF) ZM007 meta-dacite Chongwe River CH6 banded mafic gneiss Chowe River CH7 meta-tuff Chowe River CH7 meta-tuff Chowe River CH9 meta-dacite Chowe River CH9 meta-dacite Chowe River CH10 K-feldspar augen gneiss Chowe River CH10 K-feldspar augen gneiss Chowe River Charnockite associated with Chipera gabbro–anorthosite Garnet –spinel–cordierite gneiss (Chipata Gneiss) Porphyritic granite (EP26 Petauke–Sinda Terrane) Deformed K-feldspar augen gneiss CHP2a Undeformed syenite CHP2c Moderately deformed coarsegrained syenite CHP3 Opx-bearing granulite from Madzimoyo quarry CHP4a Garnet –opx-bearing mafic layer Madzimoyo quarry CHP4b Opx-bearing granulite roadside Madzimoyo quarry CHP5 Coarse-grained hbl–biotite equigranular granite CHP6a Coarse-grained K-feldspar porphyritic granite CHP8 Coarse-grained K-feldspar porphyritic granite CHP10 Foliated or banded qtz–feldspar migmatite, leucosome portion CHP11a K-feldspar porphyritic granite CHP12 K-feldspar porphyritic granite CHP13 Magmatically banded K-feldspar porphyritic granite PS17 Coarse-grained bt-poor qtz–plag granite PS18
Method
Age (Ma) Error
Block
Source
SHRIMP
664
27
S. Malawi
Kro¨ner et al. (2001)
SHRIMP SHRIMP
576 1393
11 22
S. Malawi S. Irumide Belt
Kro¨ner et al. (2001) Johnson et al. (2005)
SHRIMP
1082
7
S. Irumide Belt
Johnson et al. (2005)
SHRIMP SHRIMP
1066 1071
21 8
S. Irumide Belt S. Irumide Belt
Johnson et al. (2005) Johnson et al. (2005)
SHRIMP
1083
8
S. Irumide Belt
Johnson et al. (2005)
SHRIMP
1088
20
S. Irumide Belt
Johnson et al. (2005)
SHRIMP
1051
12
S. Irumide Belt
Johnson et al. (2005)
SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP
1064 1037 1040 1105 1094
15 8 21 22 2
S. Irumide Belt S. Irumide Belt S. Irumide Belt S. Irumide Belt S. Irumide Belt
Johnson et Johnson et Johnson et Johnson et Johnson et
SHRIMP
1105
9
S. Irumide Belt
Johnson et al. (2005)
TIMS
1050
20
S. Irumide Belt
Johnson et al. (2005)
TIMS
1046
3
S. Irumide Belt
Johnson et al. (2005)
LA-ICP-MS
1125
15
S. Irumide Belt
Johnson et al. (2005)
SHRIMP
1046
4
S. Irumide Belt
Johnson et al. (2006)
SHRIMP SHRIMP
1050 543
7 6
S. Irumide Belt S. Irumide Belt
Johnson et al. (2006) Johnson et al. (2006)
SHRIMP
1076
6
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
1977
11
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
1047
20
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
1038
6
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
1061
13
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
1076
14
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
1950
67
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
1038
9
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
1058
34
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
479
9
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
510
6
S. Irumide Belt
Johnson et al. (2006)
al. (2005) al. (2005) al. (2005) al. (2005) al. (2005)
(Continued)
ANTARCTIC–MOZAMBIQUE –SRI LANKA CORRELATION
99
Table 1. Continued Unit and sample number Fine-grained magmatically banded syenite PS19 Medium-grained, equigranular qtz–plag syenite PS 28 Undeformed equigranular coarse qtz–Kfs –bt granite PS65 Undeformed K-feldspar porphyritic granite PS71b Patch equigranular granite in coarse pegmatites PS73 Equigranular medium-grained qtz–plag –bt granite PS76 Strongly deformed quartzofeldspathic gneiss PS78 Garnet-bearing pelitic migmatite SZ16 Deformed qtz–feld gneiss with thin amphibolite SZ23 Progressively mylonitized porphyritic granite SZ25c Strongly deformed quartzofeldspathic gneiss SZ26 Strongly deformed hornblende – biotite gneiss SZ27 Tonalite Angonia Complex GGZ238 Metabasite Angonia Complex GGZ229 Monzonite GGZ256 Desaranhama Granite Monte Capingo Suite Sinda granite Monte Capirimpica Suite Cassacatiza Suite Monte Sanje Suite Granito Castanho Chipera Complex (Tete Suite) Macanga Granite Mussata Granite Ocua Complex Charnockite Marrupa Complex Tonalitic Gneiss Granitic gneiss (MA1) Diorite granulitic gneiss (MA2) Trondhjemite gneiss (MA3) Quartz monzonite gneiss (MA4) Charnockitic gneiss (MA6) Charnockitic gneiss(MA7) Pelitic paragneiss (MA8) Charnoenderbitic gneiss (MA9) Biotite–hornblende gneiss (MA10) Biotite–hornblende gneiss (MA13) Biotite–hornblende gneiss (MA13) Biotite–hornblende gneiss (MA14)
Method
Age (Ma) Error
Block
Source
SHRIMP
494
5
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
495
10
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
1043
14
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
720
12
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
504
7
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
474
8
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
742
13
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
1984
21
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
1008
17
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
1023
12
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
1961
31
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
647
11
S. Irumide Belt
Johnson et al. (2006)
SHRIMP
1104
11
S. Irumide Belt
Grantham et al. (2007a)
SHRIMP
1058
11
S. Irumide Belt
Grantham et al. (2007a)
SHRIMP
SHRIMP SHRIMP
568 1041 1201 502 1086 1077 1050 1050 1047 470 1046 994 951
5 4 10 8 7 2 8 2 29 14 20 61 44
S. Irumide Belt S. Irumide Belt S. Irumide Belt S. Irumide Belt S. Irumide Belt S. Irumide Belt S. Irumide Belt S. Irumide Belt S. Irumide Belt S. Irumide Belt S. Irumide Belt N. of Lurio N. of Lurio
Grantham et al. (2007a) Ma¨ntta¨rri et al. (2006) Ma¨ntta¨rri et al. (2006) Ma¨ntta¨rri et al. (2006) Ma¨ntta¨rri et al. (2006) Ma¨ntta¨rri et al. (2006) Ma¨ntta¨rri et al. (2006) Ma¨ntta¨rri et al. (2006) Ma¨ntta¨rri et al. (2006) Ma¨ntta¨rri et al. (2006) Ma¨ntta¨rri et al. (2006) Macey et al. (2007) Grantham et al (2007b)
Pb/Pb Pb/Pb Pb/Pb Pb/Pb Pb/Pb Pb/Pb Pb/Pb Pb/Pb Pb/Pb
evap. evap. evap. evap. evap. evap. evap. evap. evap.
602.7 644.9 582.9 577.5 590.5 928.9 576.7 1012.5 998.9
1 0.9 1 1 1 0.9 1 0.8 0.8
Southern Southern Southern Southern Southern Southern Southern Southern Southern
Kro¨ner et al. Kro¨ner et al. Kro¨ner et al. Kro¨ner et al. Kro¨ner et al. Kro¨ner et al. Kro¨ner et al. Kro¨ner et al. Kro¨ner et al.
Pb/Pb evap.
738.7
0.9
Southern Malawi Kro¨ner et al. (2001)
Pb/Pb evap.
576.1
1
Southern Malawi Kro¨ner et al. (2001)
Pb/Pb evap.
1040.6
0.7
Southern Malawi Kro¨ner et al. (2001)
Sm –Nd
Malawi Malawi Malawi Malawi Malawi Malawi Malawi Malawi Malawi
(2001) (2001) (2001) (2001) (2001) (2001) (2001) (2001) (2001)
(Continued)
100
G. H. GRANTHAM ET AL.
Table 1. Continued Unit and sample number
Method
Charnoenderbitic gneiss (MA15) Biotite–hornblende gneiss (MA16) Biotite gneiss (MA17) Biotite gneiss (MA17) Biotite gneiss (MA17)
Age (Ma) Error
Block
Source
Pb/Pb evap. Pb/Pb evap.
554.7 667.5
1 0.9
Southern Malawi Kro¨ner et al. (2001) Southern Malawi Kro¨ner et al. (2001)
Pb/Pb evap. Pb/Pb evap. Pb/Pb evap.
710.5 556.1 772.5
0.9 1 0.5
Southern Malawi Kro¨ner et al. (2001) Southern Malawi Kro¨ner et al. (2001) Southern Malawi Kro¨ner et al. (2001)
The data from north of the Lurio Belt are from Jamal (2005) and Bingen et al. (2006).
Table 2. Crystallization ages from the Nampula Block Rock type and sample number Granulite Migmatitic granite gneiss (sample MS5) Leucocratic granite (sample MS6) Augen gneiss sample NHF Tonalitic gneiss sample CVGN Granite Megacrystic charnockite Augen gneiss Tonalite Mocuba Suite Augen gneiss Mocuba Gneiss Augen gneiss Augen gneiss Tonalitic gneiss Augen gneiss Granitic gneiss Calc-silicate Granite Granitic gneiss Granite Augen gneiss Tonalitic gneiss Granite Syenite Granite Granite Granite NB21-8 Granite NB21-1 Granite NB5-1 Syenite Granulite
Unit Mocuba Complex
Nhansipfhe Megacrystic Gneiss Chimoio Granodiorite Gneiss Murrupula Suite Culicui Suite Culicui Suite Mocuba Complex Mocuba Complex Culicui Suite Mocuba Complex Culicui Suite Culicui Suite Rapale gneiss Culicui Suite Mamala Gneis Molucue Grp Murrupula Suite Molucue Grp Murrupula Suite Culicui Suite Rapale gneiss Murrupula Suite Murrupula Suite Murrupula Suite Murrupula Suite Murrupula Suite Murrupula Suite Murrupula Suite Ramiane Suite Monapo Complex
Additional data are available from Jamal (2005).
Method
Age (Ma)
Error
Source
SHRIMP SHRIMP
1028 1094
7 13
Costa et al. (1994) Kro¨ner et al. (2001)
SHRIMP
1009
13
Kro¨ner et al. (2001)
SHRIMP
1112
18
Manhica et al. (2001)
SHRIMP
1108
12
Manhica et al. (2001)
SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP
495 1074 1082 1078 1123 1085 1129 1077 1092 1091 1076 1092 1127 533 1090 504 1073 1095 521 527 507 497 516 505 514 634 637
2 13 26 16 14 10 9 26 42 14 8 13 11 5 22 12 16 8 4 4 7 4 3 5 4 8 6
Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Macey et al. (2007) Grantham et al. (2007b) Grantham et al. (2007b)
ANTARCTIC–MOZAMBIQUE –SRI LANKA CORRELATION
(Barton et al. 1993; Dirks et al. 1998; Vinyu et al. 1999; Hargrove et al. 2003; Fig. 2) as well as the Guro Bimodal Suite (Ma¨ntta¨ri et al. 2006; Westerhof 2006). These rocks are located along the margin of the Zimbabwe Craton and consequently their relationship to the Mozambique Belt is uncertain. Koistinen et al. (2006) related the c. 850 Ma magmatism to extensional processes at the margin of the Zimbabwe Craton, whereas Westerhof (2006) suggested that these ages may be related to detachment thrusting. The c. 850 Ma ages are therefore geographically restricted to areas at or close to the Zimbabwe Craton and consequently are anomalous in the Nampula Block. The difference in ages between the Namuno and Nampula Blocks was recognized by Pinna (1995), although at that time the nature and origin of the age differences was unclear. Another important difference between the three areas is that the Nampula Block and southern Irumide area have some samples with crystallization ages between 1100 and 1200 Ma whereas these ages are absent in the north of the Lurio Namuno area. Another difference is that the southern Irumide area appears to have dominantly 1000–1100 Ma ages and fewer ages in the ,600 Ma range. It is uncertain whether this is real or an artefact of sampling. Metamorphic age data are summarized in Tables 3 (Namuno Block) and 4 (Nampula Block). Data sources include Kro¨ner et al. (1997, 2001), Jamal (2005), Johnson et al. (2005, 2006), Bingen et al. (2006), Grantham et al. (2007a, b), and Macey et al. (2007). Comparison of the metamorphic ages shows that from the Namuno Block (Fig. 3c) no evidence of Mesoproterozoic metamorphism has been recorded north of the Lurio Belt, with metamorphic ages in the Namuno Block being c. 750– 400 Ma with 1050–1100 Ma metamorphism being recorded in the southern Irumide area. In contrast, data from the Nampula Block (Fig. 3d) indicate that metamorphism occurred during the Mesoproterozoic between 1050 and 1100 Ma as well as during the time period c. 600–400 Ma. In conclusion, from the histograms shown in Figure 3, it is recognized that the Namuno and Nampula Blocks have different characteristics, some fairly distinct (e.g. the common presence of 600 –900 Ma ages in the Namuno Block and absence in the Nampula Block) and some subtle.
Antarctica Extending the patterns recognized from the Namuno and Nampula blocks to the adjacent areas of Dronning Maud Land (DML), Antarctica, in a Gondwana context (Fig. 4), the following aspects become apparent. The age distribution for the Nampula Block is virtually identical to that
101
observed in western DML (Sverdrupfjella þ Kirwanveggean) (Fig. 5a) with crystallization ages in both areas falling in the time periods 950 – 1200 Ma and 470 – 500 Ma and metamorphic ages being recorded for the periods 950 – 1100 Ma and 450 – 600 Ma (Fig. 5b). The data utilized in Figure 5a and b are summarized in Tables 5 and 6 with the data derived from Harris et al. (1995), Krynauw & Jackson (1996), Jackson & Armstrong (1997), Harris (1999), Jackson (1999), Board et al. (2005), Grantham et al. (2006) and G. H. Grantham & R. A. Armstrong (unpubl. data). In addition, equivalents of the megacrystic granitic augen gneiss Culicui Suite and heterogeneous medium-grained, equigranular tonalitic orthogneiss Mocuba Suite, both volumetrically significant lithological units in the Nampula Block, are recognized in western DML in the form of the Kirwanveggen megacrystic orthogneiss (Grantham et al. 1995) and the Kvervelknatten orthogneiss (Grantham et al. 1995, 1997; Groenewald et al. 1995; Wareham et al. 1998), respectively. The chemistries of these two units are comparable, as are the ages, which are typically c. 1070–1090 and c. 1110– 1140 Ma, respectively. Progressing further eastwards into central DML (excluding Schirmacher Hills), the western Mu¨hlig-Hofmannfjella has crystallization ages and metamorphic ages largely comparable with those of the Nampula and western DML areas of Antarctica (Fig. 5c and d). The data from central DML are summarized in Tables 7 and 8, with the data sources including Jacobs et al. (1998, 2003a–c), Paulsson & Austrheim (2003), Bisnath et al. (2006). The crystallization ages recorded for central DML are in the range of 450–650 Ma and 1050– 1200 Ma and the metamorphic ages 500–600 and 1000–1100 Ma. The 550–650 Ma crystallization ages from central DML are all collected from the extreme eastern end of central DML from charnockites and anorthosites in the Wolthaat Massif area and are not recognized in the western Mu¨hlig-Hofman Mountains. Charnockites and anorthosites of Neoproterozoic age are not recognized in the Nampula Block. In contrast, limited data from the granulites exposed in Schirmacher Hills in NE Mu¨hligHofmannfjella (Table 8) dominantly have ages in the range c. 550–700 Ma with a few older ages between 800 and 1150 Ma being recognized (Ravikant et al. 2004, 2008). Consequently, the Schirmacher Hills has ages comparable with those of the north of the Lurio Namuno Block (Fig. 5e). Further east in the Sør Rondane area of Antarctica, SHRIMP (Shiraishi et al. 2008) and chemical Th –U –total Pb isochron method (CHIME) data (Asami et al. 2005) indicate that the NE Sør Rondane has age distributions similar
102
G. H. GRANTHAM ET AL.
Table 3. Metamorphic ages used for the Namuno Block Rock type and sample number
Method
Age (Ma)
Error
Area
Source
Charnockitic gneiss (MA8) Felsic granulite (MA12) Biotite–hornblende gneiss (MA13) Hofineir Gneiss deformed quartzofeldspathic gneiss Garnet-bearing pelitic migmatite Nyamadzi Gneiss deformed quartzofeldspathic gneiss Wutepo Gneiss deformed hornblende –biotite gneiss Titanite in mafic gneiss Ocua Complex MZ05045a charnockite Mugeba Complex
SHRIMP SHRIMP SHRIMP
572 547 564
9 10 4
S. Malawi S. Malawi S. Malawi
Johnson et al. (2005) Johnson et al. (2005) Johnson et al. (2005)
SHRIMP
536
10
S. Irumide area
Johnson et al. (2006)
SHRIMP SHRIMP
1942 1065
5 13
S. Irumide area S. Irumide area
Johnson et al. (2006) Johnson et al. (2006)
SHRIMP
555
11
S. Irumide area
Johnson et al. (2006)
SHRIMP SHRIMP
549 555
7 5
S. Irumide area N. of Lurio
Grantham et al. (2007a) Macey et al. (2007)
SHRIMP
614
8
Mugeba
Kro¨ner et al. (2001)
Additional data are from Jamal (2005) and Bingen et al. (2006).
to those of the Namuno Block (Fig. 5f). Almost all of these samples are from the NE Sør Rondane. The SW Sør Rondane is separated from the NE Sør Rondane by a c. 10 km wide shear zone (Shiraishi et al. 1991; Shiraishi & Kagami 1992). The crystallization ages from Sør Rondane range between 1200 and 500 Ma with most being between 500 and 650 Ma whereas metamorphic ages are between
500 and 650 Ma with no Mesoproterozoic metamorphism being recognized.
Sri Lanka Very few SHRIMP zircon U/Pb or single-grain zircon data from single suites are available from Sri Lanka. The available data (Kro¨ner et al. 1987,
Table 4. Metamorphic ages from the Nampula Block Unit Mugeba Complex Mocuba Suite Mocuba Suite Molocue Group Culicui Suite Culicui Suite Culicui Suite Rapale Gneiss Culicui Suite Mamala Fm. Leucosome S3 in Culicui Suite Mocuba Suite Rapale Gneiss Monapo Complex Monapo Complex
Rock type
Method
Granulite zircon rim Quartzofeldspathic gneiss zircon rim Migmatitic vein Quartzofeldspathic gneiss zircon rim Augen gneiss zircon rim Augen gneiss zircon rim Charnockite zircon rim Tonalitic gneiss lower intercept Augen gneiss zircon rim Quartzofeldspathic gneiss zircon rim Leucosome
SHRIMP SHRIMP
591 527
4 18
Macey et al. (2007) Macey et al. (2007)
SHRIMP SHRIMP
1063 502
47 90
Macey et al. (2007) Macey et al. (2007)
SHRIMP SHRIMP SHRIMP SHRIMP
538 525 513 449
8 20 10 95
Macey et al. Macey et al. Macey et al. Macey et al.
SHRIMP SHRIMP
505 555
10 12
Macey et al. (2007) Macey et al. (2007)
SHRIMP
490
8
Macey et al. (2007)
Tonalitic gneiss zircon rim Tonalitic gneiss lower intercept Granulite zircon rim Syenite zircon rim
SHRIMP
1090
34
Macey et al. (2007)
SHRIMP
608
42
Macey et al. (2007)
SHRIMP SHRIMP
579 596
11 5
Grantham et al. (2007b) Grantham et al. (2007b)
Additional data are available from Jamal (2005).
Age (Ma) Error
Source
(2007) (2007) (2007) (2007)
ANTARCTIC–MOZAMBIQUE –SRI LANKA CORRELATION
103
Fig. 4. Schematic map of the various crustal blocks belonging to the Namuno and Nampula-type blocks. VC, Vijayan Complex; CHC, Central Highland Complex; Mo, Monapo Klippen; Mg, Mugeba Klippen; Kg, Kataragama Klippen; L, Lurio Belt; SH, Schirmacher Hills; MH, Mu¨hlig-Hofmannfjella; GV, Gjelsvikfjella; HUS, Sverdrupfjella; Kvg, Kirwanveggen; Hmf, Heimefrontfjella; GC, Grunehogna Craton; Ur, Urungwe Klippen; PCM, Prince Charles Mountain.
1994; Baur et al. 1991; Holzl et al. 1994) are largely derived from multigrain TIMS studies on single rock units or come from SHRIMP studies on metasediments aimed at provenance determinations. None the less, histograms summarizing data from the Highland Complex (HC) (Fig. 5g) and Vijayan Complex (VC) (Fig. 5h) of Sri Lanka show that the available data from the HC have a similar pattern to those for the Namuno Block of Mozambique and the Sør Rondane and Schirmacher Hills areas of Antarctica, whereas the VC has a pattern of ages comparable to those of the Nampula Block and DML, Antarctica. The data reported from the Vijayan Complex by Kro¨ner et al. (1987) have large analytical uncertainties, resulting in the broad curves defined by the probability density distribution (Fig. 5h), whereas the absolute ages show age ranges of 500– 600 and 1000– 1250 Ma. In conclusion, the chronological data combined with lithological varieties facilitate recognition of two age groups of tectonic blocks; namely, those with significant volumes of rock with ages
between 600 and 900 Ma and those without (Fig. 4). The former group comprises the north of the Lurio Namuno Block, Malawi and southern Irumide Block as well as the Mugeba and Monapo klippen, the Highland Complex in Sri Lanka, the NE Sør Rondane, the far eastern Mu¨hlig-Hofmannfjella and the Schirmacher Hills. The latter group comprises the Nampula Block, the Vijayan Complex in Sri Lanka, the SW Sør Rondane, the western Mu¨hlig-Hofmannfjella, Sverdrupfjella and its extensions into Kirwanveggen (Fig. 4). In the remainder of the paper, we will refer to these grouped blocks as the Namuno and Nampula Blocks, respectively.
Structural data In most cases the boundaries between the Namuno and Nampula Blocks (when exposed) are defined by highly sheared rocks. In NE Mozambique the boundary is represented by the highly sheared Lurio Belt and the circumferential mylonites
104
G. H. GRANTHAM ET AL.
Fig. 5. Histograms of geochronological data from western DML (a, b), central DML (c, d), Schirmacher Hills (e), Sør Rondane (f), Sri Lanka (Highland Complex) (g) and Sri Lanka (Vijayan Complex) (h). The 600 Ma and 1100 Ma lines are shown for reference. The data from Sør Rondane are subdivided into SHRIMP-based ages of crystallization and metamorphism and CHIME-based ages of crystallization and metamorphism.
ANTARCTIC–MOZAMBIQUE –SRI LANKA CORRELATION
105
Table 5. Crystallization ages from Sverdrupfjella and Kirwanveggen, western Dronning Maud Land Unit
Method
Age (Ma)
Error
Kyanite leucogneiss Bt–grt migmatite Intrusive leucogneiss Megacrystic orthogneiss Pegmatite vein Late felsic dyke Sveabreen Granite Fugitive Granite Fugitive Granite Roerkulten Granite Rootshorga Paragneiss Granite gneiss tonalitic Wbsv065 Tabular granite Wbsv073 Granite dykes Wbsv069 Kvervelkatten Gneiss Kvervelkatten Amphibolite Pod Cjk151 Megacrystic augen gneiss Cjk158 Leucopegmatite Cjk155 Porphyritic granite dyke Cjk 103 Leucogranite Cjk152 Granite dyke Cjk 159 Amphibolite dyke Hallgrens Cjk160 Banded bt gneiss Hallgrens Cjk56 Banded bt gneiss Hallgrens Cjk59 Titanite Cjk 159 Grey gneiss Dalmatian Granite Brattskarvet Monzonite Midbressrabben Diorite
SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP
1096 1157 1101 1088 1079 1011 1127 1131 1061 1103 1092 1132 1072 480 1134 1139 1074 1050 1011 990 980 986 1081 994 1003 1143 489 474 1140
10 10 13 10 6 8 12 25 14 13 13 16 10 10 11 10 11 10 8 12 13 6 4 22 9 11 10 10 10
Grey gneiss Jutulrora Jw4
SHRIMP
1139
12
Augen gneiss Sa 10
SHRIMP
1096
14
around the Monapo and Mugeba klippen. In the west the Sanangoe thrust zone (Barr & Brown 1987) and the shear zones defining the allochthonous Tete Complex (Koistinen et al. 2006; Westerhof 2006) represent possible extensions of the Lurio Belt. In Sri Lanka, the boundary between the Highland Complex is interpreted as a complex thrustfault zone (Kleinschrodt 1994) in which the granulite-facies Highland Complex has been thrust-faulted over the amphibolite-facies Vijayan Complex. In the shear zone, which is reportedly hundreds of metres wide, a strong north–south-oriented stretching lineation is developed; however, kinematic indicators are sparse. In Sør Rondane, NE Sør Rondane is separated from SW Sør Rondane by a c. 10 km wide shear zone (Shiraishi et al. 1991). The boundary in Mu¨hlig-Hofmannfjella is, however, not exposed. The differences in reported geochronology and
Source Harris (1999) Harris (1999) Harris (1999) Harris (1999) Harris (1999) Moyes & Harris (1996) Moyes & Harris (1996) Moyes & Harris (1996) Moyes & Harris (1996) Moyes & Harris (1996) Moyes & Harris (1996) Board et al. (2005) Board et al. (2005) Board et al. (2005) Jackson (1999) Jackson (1999) Jackson (1999) Jackson (1999) Jackson (1999) Jackson (1999) Jackson (1999) Jackson (1999) Jackson (1999) Jackson (1999) Jackson (1999) Jackson (1999) Krynauw & Jackson (1996) Krynauw & Jackson (1996) G. H. Grantham & R. A. Armstrong (unpubl. data) G. H. Grantham & R. A. Armstrong (unpubl. data) G. H. Grantham & R. A. Armstrong (unpubl. data)
lithologies in eastern Mu¨hlig-Hofmannfjella imply that the boundary between the two blocks probably passes immediately east of the Wolthaat Anorthosite Massif, the most easterly nunatak group in Mu¨hlig-Hofmannfjella, whose age and extensional affinity suggest that it belongs to the Namuno Block along with the granulites at Schirmacher Hills and Mramornye nunataks. This is possibly also supported by the differences in structural histories between the Wolthaat Massif and rocks further east described by Bauer et al. (2004). Immediately east of the Wolthaat Massif, the nunataks are reportedly underlain by c. 550 Ma granites after which, progressing eastwards, the lithologies are typical of the Nampula Block (Jacobs et al. 2003c; Bauer et al. 2004). The lithologies astride the Orvinfjella shear zone in Mu¨hlig-Hofmannfjella reportedly are not different (Jacobs et al. 2003c; Bauer et al. 2004) and the orientation of the Orvinfjella Shear Zone suggests that it may
106
G. H. GRANTHAM ET AL.
Table 6. Metamorphic ages from Sverdrupfjella and Kirwanveggen, western Dronning Maud Land Unit
Method
Age (Ma)
Error
Source
Leucosome in garnet migmatite gneiss Kyanite leucogneiss Granite gneiss rim tonalitic wbsv065 Metapelite wbsv025 Metapelite wbsv025 Tabular granite rim wbsv073 Tabular granite rim wbsv073 Leucosome wbsv071 Leucosome wbsv071 lower intercept Leucosome wbsv074 Leucosome wbsv113 Leucosome wbsv113 Leucosome wbsv114 Leucosome wbsv116 Rim Kvervelknatten
SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP
1098 1096 1031 1044 540 565 996 1035 499 515 1032 503 525 519 1060
5 10 47 47 6 11 17 31 17 7 15 35 35 4 22
Leucosome CJK153 Monazite CJK 149 Titanite CJK56 Rim SA10
SHRIMP SHRIMP SHRIMP SHRIMP
1031 956 1015 538
6 17 16 25
Mafic dyke RK55
SHRIMP
523
21
Harris (1999) Harris (1999) Board et al. (2005) Board et al. (2005) Board et al. (2005) Board et al. (2005) Board et al. (2005) Board et al. (2005) Board et al. (2005) Board et al. (2005) Board et al. (2005) Board et al. (2005) Board et al. (2005) Board et al. (2005) Krynauw & Jackson (1996) Jackson (1999) Jackson (1999) Jackson (1999) G. H. Grantham & R. A. Armstrong (unpubl. Data) Grantham et al. (2006)
continue into Mozambique as the Namama ShearZone (Grantham et al. 2003), where it dissects similar rock types. Piazolo (2004) has described the structural evolution of the Mramornye nunataks and Schirmacher Hills. Both areas are characterized by strong shear fabrics, with those at Mramornye nunataks suggesting thrust-faulting toward the south during D3.
Planar fabric data Grantham et al. (2003) described one of the enigmas of the correlation of northern Mozambique with Dronning Maud Land as being the opposing structural facing directions in the two areas. Planar structures in northern Mozambique were described as dipping dominantly to the north and NW whereas those in western Dronning Maud Land were described as dipping dominantly to the SE. The improved densities of structural observation and geochronology permit the conclusion that the ages of fabrics in the two areas are not the same. The strong planar NW-dipping fabrics in and adjacent to the Lurio Belt clearly affect rocks with crystallization ages of c. 630 Ma along with older rocks in the Nampula Block. Consequently, the strong fabric-producing event in these areas has to be younger than c. 630 Ma. The structural data from the Mugeba and Monapo klippen, particularly
the latter, are discordant to their structurally underlying rocks, supporting their interpretation as klippen (Grantham et al. 2007b; Macey et al. 2007). The Monapo Complex, in particular, is interpreted as a circular synformal remnant in which at least two phases of deformation defined by the layered granulites form a large type 2 interference fold structure (Grantham et al. 2007b). In contrast to the younger than 630 Ma deformation in the klippen and the Lurio Belt, a detailed study in southern Kirwanveggen by Jackson (1999) showed that most of the deformation occurred there before c. 900 Ma. This is circumstantially supported in southern Kirwanveggen at Skappelknabben, where strongly sheared augen gneisses with greenschist-grade, planar fabrics are in relative close proximity to the virtually undeformed (brittlefaulted) sandstones of the Urfjell Group at Drapane. The Urfjell Group is less than c. 550 Ma old (Moyes et al. 1997), the age of the youngest detrital zircon recorded in it (Croaker 1999). Grantham et al. (2006) have shown that a c. 950 Ma mafic dyke at Roerkulten in Sverdrupfjella post-dates an earlier migmatitic fabric-forming event (D1), has a planar fabric (D2) that has been deformed about D3 folds, and was metamorphosed in the upper amphibolite facies at c. 500 Ma. Similarly, the syntectonic emplacement of granitic sheets at c. 480 Ma during top-to-the-SE-directed deformation in NW Sverdrupfjella (Grantham et al. 1991) demonstrates
ANTARCTIC–MOZAMBIQUE –SRI LANKA CORRELATION
107
Table 7. Crystallization ages for Mu¨hlig-Hofmannfjella in central Dronning Maud Land Rock unit or type and sample number
Method
Age (Ma)
Error
Source
Stabben gabbro Granite aplite dykes Augen gneiss Grey migmatite augen gneiss Granite gneiss Banded gneiss Granite gneiss Homogeneous migmatite Stabben syenite Lamprophyre dyke Risemedet/2312/2 Charnockite Hochlinfjellet 1301/2 Augen gneiss 2412/4 Grey migm gneiss 2712/4 Augen gneiss 1512/1 Grey gneiss 1701/2 Migmatitic augen gneiss Stabben gabbro Granite dyke Gygra Felsic gneiss j1671 Felsic gneiss j1704 Felsic gneiss j1795 Orthogneiss j1736 Orthogneiss j1797 Metagranodiorite j1698 Metaleucogranite j1695 Felsic gneiss j1838 Charnockite j1886 Anorthosite j1955 Anorthosite j1958 Zwiesel gabbro Zwiesel gabbro
SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SIMS SIMS SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP
487 497 1104 1124 1133 1091 1130 1163 500 523 521 1096 1115 1123 1142 1137 483 487 1073 1137 1076 1086 1087 530 527 1130 608 600 583 527 521
4 5 8 11 16 16 19 6 8 5 3 8 12 21 21 14 11 4 9 21 14 20 28 8 6 12 9 12 7 5 6
Bisnath et al. (2006) Bisnath et al. (2006) Bisnath et al. (2006) Bisnath et al. (2006) Bisnath et al. (2006) Bisnath et al. (2006) Bisnath et al. (2006) Paulsson & Austrheim (2003) Paulsson & Austrheim (2003) Jacobs et al. (2003a) Jacobs et al. (2003a) Jacobs et al. (2003b) Jacobs et al. (2003b) Jacobs et al. (2003b) Jacobs et al. (2003b) Jacobs et al. (2003b) Jacobs et al. (2003a) Jacobs et al. (2003a) Jacobs et al. (1998) Jacobs et al. (1998) Jacobs et al. (1998) Jacobs et al. (1998) Jacobs et al. (1998) Jacobs et al. (1998) Jacobs et al. (1998) Jacobs et al. (1998) Jacobs et al. (1998) Jacobs et al. (1998) Jacobs et al. (1998) Jacobs et al. (2003c) Jacobs et al. (2003c)
a direction of deformation similar to that in Mozambique. These data support the work of Grantham et al. (1995), who concluded that there were two major periods of deformation in Sverdrupfjella in western DML; namely, one during the Mesoproterozoic at c. 900 –1000 Ma and the other during the Neoproterozoic and into the Cambrian at c. 550 –490 Ma. The data do not support the suggestion by Board et al. (2005) that the dominant deformation in Sverdrupfjella is Neoproterozoic to Cambrian in age. A more detailed study of planar structures (Figs 6–8) in the various areas of Mozambique and DML, Antarctica provides a better understanding of the structural relationships. In northern Mozambique, planar fabrics (Fig. 6a and b) dip and plunge dominantly toward the north to NW in the Lurio Belt as well as in the area along the southern margin of the Lurio Belt. Similarly the lineations in the same areas (Fig. 6e) define an arc with lineations plunging between north and west with concentrations towards the north and WNW. Limited fold-axis data are similar to the lineations
shown in Figure 6e. Progressing southwards toward the northern Mozambique coast the structural patterns become more complex and bimodal in nature, with both north- to NW-dipping planar structures as well as south- to SE-dipping structures being common (Fig. 6c and d). Similarly, the lineations along the southern margin of the northern Mozambique coast also show a bimodal variation with west- and ESE-dipping orientations (Fig. 6f). The data in Figure 5f are skewed by the high number of readings collected in the broad vicinity of the Namama shear zone by Aquater in the early 1980s (Aquater 1983). Bimodal patterns in the planar fabrics are observed in the western Mu¨hlig-Hofman Mountains of Antarctica (Fig. 7f; data from Jacobs et al. 2003a) and the Gjelsvikfjella (Fig. 7g, A. Bisnath unpubl. data), with gneisses dipping broadly to the NE and SW. In contrast, the structural data from Mramornye nunataks (Fig. 7h; from Piazolo 2004) is unimodal, with the gneisses dipping dominantly shallowly to steeply toward the ENE. Lineations from Gjelsvikfjella (Fig. 7a) show at least three
108
G. H. GRANTHAM ET AL.
Table 8. Metamorphic ages from Mu¨hlig-Hofmannfjella and Schirmacher Hills Subject or sample and sample number Age of migmatization Zircon overgrowths Migmatite gneiss Lower intercept Aba/32 Leucosome-1301/2 Leucosome-0801/3 Grey gneiss 1701/2 Grey gneiss 1701/2 Augen gneiss 1512/1 Leucosome Rim j1704 Rim j1795 J1886 charnockite Anorthosite rim j1955 Metamorphic rim j1704 Whole-rock minerals Whole-rock minerals Whole-rock minerals Monazite Monazite Titanite Monazite Monazite Monazite Monazite Monazite Monazite Monazite Titanite Titanite
Method
Age (Ma)
Error
Source
SHRIMP SIMS SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP SHRIMP TIMS Sm –Nd TIMS Sm –Nd TIMS Sm –Nd TIMS U/Pb TIMS U/Pb TIMS U/Pb TIMS U/Pb TIMS U/Pb TIMS U/Pb TIMS U/Pb TIMS U/Pb TIMS U/Pb TIMS U/Pb TIMS U/Pb TIMS U/Pb
504 504 529 527 521 558 1061 528 1049 516 522 557 544 555 1084 616 632 554 629 639 580 809 656 676 580 613 1044 920 589 1146
4 6 4 50 3 6 56 10 19 5 10 11 15 11 8 52 8 16 5 4 5 6 2 12 5 4 25 5 9 3
Paulsson & Austrheim (2003) Paulsson & Austrheim (2003) Bisnath et al. (2006) Bisnath et al. (2006) Jacobs et al. (2003a) Jacobs et al. (2003a) Jacobs et al. (2003b) Jacobs et al. (2003b) Jacobs et al. (2003b) Jacobs et al. (1998) Jacobs et al. (1998) Jacobs et al. (1998) Jacobs et al. (1998) Jacobs et al. (1998) Jacobs et al. (1998) Ravikant et al. (2004) Ravikant et al. (2004) Ravikant et al. (2004) Ravikant et al. (2008) Ravikant et al. (2008) Ravikant et al. (2008) Ravikant et al. (2008) Ravikant et al. (2008) Ravikant et al. (2008) Ravikant et al. (2008) Ravikant et al. (2008) Ravikant et al. (2008) Ravikant et al. (2008) Ravikant et al. (2008) Ravikant et al. (2008)
significant concentrations toward the NE, SE and NW. This variation either suggests numerous shearrelated lineation-producing events or reflects the folding of earlier unimodal lineations to produce multiple directions. Grantham et al. (1995) recorded bimodal planar structure dipping patterns in NW Sverdrupfjella as well as northern Kirwanveggen (Fig. 7a and b). In contrast, planar fabrics (Fig. 7c and d) in southeastern Sverdrupfjella and southern Kirwanveggen are unimodal and dip dominantly toward the SE (Grantham et al. 1995). Similarly, lineations in northern Sverdrupfjella (Fig. 8b) plunge dominantly eastwards with a subordinate roughly north-plunging group. In northern Kirwanveggen lineations plunge dominantly roughly north and south (Fig. 8c). In southeastern Sverdrupfjella a unimodal lineation direction is recognized plunging dominantly toward the SE (Fig. 8d). The data suggest a zone along the southern coast of northern Mozambique and along the northern coast of DML in which bimodal structural patterns are seen in planar and linear structures. In NW
Sverdrupfjella it is apparent in the field from the wonderful 3D exposures available in Antarctica that the bimodal pattern arises from the refolding of earlier D1 and D2 SE-dipping planar fabrics about near-horizontal NE-oriented D3 fold axes (Fig. 9a and b). The D3 folds are relatively open and commonly verge toward the SE or have top-to-the-SE geometries. In contrast, the D1 and D2 folds are tight to isoclinal and commonly verge toward the NW (Fig. 9a). The folded dyke at Roerkulten reported by Grantham et al. (2006) is also typical of a D3 fold. Other examples of D3 folds with NW-dipping axial planes are seen at the southern end of Brekkerista and the western end of Roerkulten (Grantham 1992).
Metamorphic history In Mozambique the mineral assemblages and grades of metamorphism also show significant differences. Except for the Mugeba and Monapo klippen, the grade of metamorphism in the Nampula Block
ANTARCTIC–MOZAMBIQUE –SRI LANKA CORRELATION
109
Fig. 6. Contoured stereographic projections of poles to planar structures (a–d) and stereographic projections of contoured lineations (e, f) from various areas mapped in northern Mozambique. The approximate position of the Lurio Belt is shown. The arrows connecting the stereonets to the map area borders show the approximate areas from which the data have been collected.
is typically upper amphibolite facies, with orthopyroxene being seen only rarely as relict grains in magmatic charnockite granitoids or as rare localized diffuse fluid-driven(?) vein charnockitization. The Mocuba Suite gneiss is extensively migmatized, locally showing both Mesoproterozoic and the Pan-African generations of migmatization. The post-Mocuba Suite gneisses preserve only the weakly developed Pan-African migmatization. A significant aspect of the Nampula Block, however, is the abundance of undeformed to weakly deformed granitoids and pegmatites whose ages vary between c. 495 and 530 Ma. Absolute pressure constraints are difficult to constrain for the Nampula Block because of the absence of rock types (meta-pelites, metabasites) with suitable mineral assemblages. The only reliable constraints are provided by sillimanite-bearing migmatitic quartzofeldspathic gneisses implying temperatures of at least c. 700 8C and
pressures ,c. 7–8 kbar. Consequently, no P–T path for the Nampula Block is presented here. However, the weak migmatization and development of granitic melts with ages of c. 495–530 Ma imply an increase in temperature probably to 650– 750 8C during this time period. Implicit in the temperature increase is an increase in depth, because no widespread extensive source of advective heat is recognized. The granitoids are concentrated in the Nampula Block, with only a limited number of small granitoids being recognized north of the Lurio Belt. In contrast, the rocks in the Mugeba and Monapo klippen contain granulite-grade orthogneisses and paragneisses. The orthogneisses are typically ultramafic, mafic to felsic in composition. The ultramafic rocks in the Monapo klippen comprise clinopyroxenites with ,c. 5% modal plagioclase but c. 20% normative plagioclase, implying a
110
G. H. GRANTHAM ET AL.
Fig. 7. Contoured stereographic projections of poles to planar structures from various localities in western and central DML.
substantial omphacitic component. Subtle vermicular intergrowths of Pl þ Cpx (c. 5% Al2O3) (mineral abbreviations after Kretz 1983) are locally developed at the margins of coarse cpx grains with c. 10% Al2O3. These are interpreted as decompression exsolution intergrowths (G. H. Grantham, unpubl. data). The mafic rocks contain Pl–Opx–Cpx–Grt, and some samples have decompression textures defined by garnet with vermicular rims of Cpx/Hbl þ Pl. In the felsic granulites idiomorphic post-tectonic garnet (þ Qtz) after Opx/Cpx define isobaric cooling reactions. The metapelites contain Grt þ Sill þ Pl þ Rt assemblages. Thermobarometry on these assemblages from Mugeba (Roberts et al. 2005) and Monapo (Grantham et al. 2007b) have facilitated the construction of P–T loops (Fig. 10b and c). The P– T loops from Mugeba and Monapo have initial isothermal decompression from c. 900 to 1000 8C and .c. 10 kbar, followed by isobaric cooling at c. 700 8C and c. 6–7 kbar. These P –T loops are comparable with those from Schirmacher Hills (Fig. 10e), Sør Rondane (Fig. 10d) and the
Highland Complex of Sri Lanka (Fig. 10a). All these P–T loops suggest early isothermal decompression followed by isobaric cooling with P– T conditions of c. 6-7 kbar and c. 600–700 8C at c. 550 Ma, although the P–T path from Sør Rondane shows additional complexities not recognized from other areas. The P– T conditions for Schirmacher Hills and Sør Rondane are from Baba et al. (2006) whereas those from Sri Lanka are from Schumacher et al. (1990), Hiroi et al. (1994) and Raase & Schenk (1994). The P–T conditions in Gjelsvikfjella are from Bisnath & Frimmel (2005; Fig. 10f) and show an isothermal decompression path from c. 10 kbar and 700–800 8C during the Mesoproterozoic toward c. 5 kbar and c. 650 8C at 550 Ma. Extensive granitoids of c. 550 Ma age are recognized in Gjelsvikfjella and Mu¨hlig-Hofmannfjella as in the Nampula terrane, also implying a thermal increase at this time. The P–T path described by Grantham et al. (1995) (Fig. 10g) is similar to that described for Gjelsvikfjella and is also characterized by
ANTARCTIC–MOZAMBIQUE –SRI LANKA CORRELATION
111
Fig. 8. Stereographic projections of lineations from various localities in western Dronning Maud Land.
granitic magmatism at c. 6 kbar and c. 700 8C at c. 490 Ma (Grantham et al. 1991). Progressing southward in Sverdrupfjella to Kirwanveggen, P– T conditions at Neuemayerskarvet in the north of Kirwanveggen (Fig. 10h) described by Grantham et al. (2001) are of the order of c. 6.5 kbar and c. 700 8C. Significantly, c. 500 Ma granitoids are absent in northern Kirwanveggen, and the P–T estimates described by Grantham et al. (2001) were based on a thermally driven dehydration reaction of Hbl þ Pl þ Qtz ! Grt þ Ab þ H2O. Progressing southwards to southern Kirwanveggen, at Drapane, c. 530 Ma sandstones and grits of the Urfjell Group are exposed, implying that at c. 530 Ma the southern Kirwanveggen was exposed at surface. These data imply a crustal depth gradient between north and south Kirwanveggen of c. 6 kbar or c. 20 km. The most important and fundamental difference between these various terranes is that in those areas with Namuno Block age signatures (Mugeba, Monapo, Highland Complex of Sri Lanka, NE Sør Rondane and Schirmacher Hills) the P –T evolution at c. 550 Ma is interpreted as involving significant isobaric cooling. In contrast, rocks with Nampula Block age signatures (Nampula, Sverdrupfjella, northern Kirwanveggen, Gjelsvikfjella and western Mu¨hlig-Hofmannfjella and southwestern Sør Rondane) are largely characterized by extensive magmatism at 500 –550 Ma, implying thermal
heating. This aspect has been recognized by Baba et al. (2008).
Discussion and conclusions This integrated study of geochronological and structural data and metamorphic P–T paths supports an interpretation that rocks north of the Lurio Belt have been thrust southwards over the Nampula (Mozambique)–Maud (Antarctica) block as summarized in Figure 11. Figure 11 represents a schematic cross-section from northern Mozambique to southern Kirwanveggen with the staggered horizontal line representing current exposure levels in Africa and Antarctica superimposed on the inferred 600–500 Ma topographic profile. At the northern end, the Geci Group sediments were deposited at c. 580 Ma (Melezhik et al. 2006; Fig. 2). Progressing southward, metamorphic conditions increase toward the Lurio Belt, with the Namuno Block being thrust over the Nampula Block. In the footwall, depression to greater depths resulted in thermal increase and partial melting, resulting in anatexis and granite genesis in the Nampula Block in Mozambique and in western DML, Antarctica. In this context, the numerical modelling of crustal melting in continental collision zones by England & Thompson (1986) is applicable; they modelled the
112
G. H. GRANTHAM ET AL.
Fig. 9. Field photographs from Jutulrora, Sverdrupfjella, western DML. The upper photograph shows a NW-vergent F1 isoclinal recumbent fold with axial planar (AP) foliation, a NW-vergent F2 isoclinal fold in which the banding is clearly folded and a small SE-vergent F3 concentric fold. The lower photograph shows a larger-scale SE-vergent F3 fold with NW-dipping axial plane. (Note also the SE-dipping thin granitic sheets in the lower photograph.)
P–T evolution in a setting where crustal thickness is doubled by large-scale thrust faulting. Their model predicts that anatexis in the footwall would result in granite genesis c. 40 Ma after the thrust-related thickening, depending on the level of anatexis. This modelling provides a plausible explanation of why the c. 530– 495 Ma granites appear to be mostly undeformed and younger than the metamorphic ages, which start at c. 590 Ma. Following the thrust-related crustal thickening, isostatic rebound with associated inversion and extensional collapse would follow, with concomitant genesis and intrusion of the granites and pegmatites. The extensional structures resulting from the collapse of the orogenic pile would be oriented at c. 908 to the s3 direction of the orogenic compression, as demonstrated in the Himalayas (Dewey 1988). Undeformed pegmatite dykes in the Nampula Block south of the Lurio Belt are correlated with the granites and have strikes
dominantly toward the NNW, implying ENE – WSW extension (Grantham et al. 2007b). The section from northern to southern Kirwanveggen represents a progression from midcrustal levels to the surface, with the deposition of the Urfjell Group at c. 530 Ma. Figure 11 also shows that the c. 550–580 Ma top-to-the-SE deformation was superimposed on an older top-tothe-NW deformation recognized in western DML. The reorientation of structures immediately below the suture zone in the footwall is shown in Figure 11 and is consistent with the structures shown in Figure 9. The interpretation of a large-scale thrust of East African Orogen rocks onto Antarctica has already been proposed by Ravikant et al. (2004, 2008), although their study did not examine the structural and metamorphic details. The interpretation presented in Figure 10 requires different levels of erosion between Africa and Sri Lanka and Antarctica. This difference in level of erosion is supported by the klippen of granulite remnants at Mugeba and Monapo on top of the Nampula Block footwall, as well as similar klippen remnants also preserved at Kataragama in Sri Lanka (Kriegsman, 1995, amongst others) (Fig. 4). Additional klippen along the northern Kalahari Craton margin that have similar geochronology (where data are available) include the Naukluft Mountains in Namibia (Ahrendt et al. 1978; Gray et al. 2006), the Urungwe Klippen in northern Zimbabwe (Shackleton et al. 1966), the Makuti Group (Dirks et al. 1999) and the allochthonous Masoso Suite (Dirks & Jelsma 2006). Within the Zambezi Belt, the main phase of deformation involved transcurrent shearing and SW-vergent thrusting (Hanson et al. 1994; Wilson et al. 1997). In contrast, Antarctica has not been subject to the same level of erosion, resulting in the preservation of a largely continuous slab of East African Orogen rocks from the nunataks at Mramornye c. 718S, c. 88E (Piazolo 2004) via Schirmacher Oasis (Ravikant et al. 2002, 2004, 2008; Baba et al. 2006, 2008), the eastern Mu¨hlig-Hofman mountains, NE Sør Rondane and beyond. We have not explored detailed aspects of the geology east of Sør Rondane other than to note that we are not aware of any rocks, east of SW Sør Rondane, with Nampula Block type lithological, geochronological and metamorphic characteristics. This may suggest that it is possible that rocks of the Nampula Terrane do not extend beyond the SW Sør Rondane. Using the broad geographical grid references in Figure 4, the width of the overthrust block approximates to 4–58 of latitude, equivalent to c. 480– 600 km, representing the distance from the Lurio Belt in Mozambique to the sheared boundary
ANTARCTIC–MOZAMBIQUE –SRI LANKA CORRELATION
113
Fig. 10. Figure summarizing P– T loops from various localities in Mozambique, Sri Lanka, central DML, western DML and Sør Rondane as shown by the arrows linking the P –T loops with the geographical locality. Abbrevations as in Figure 4.
between the two crustal blocks in Sør Rondane and Mu¨hlig-Hofmannfjella. At the western extremity of the proposed belt, Martin (1974) estimated that the c. 560 –570 Ma (Gray et al. 2006) Naukluft nappes had been transported at least 60 km towards the SE onto the Kalahari Craton. For purposes of comparison, it should be noted that recent studies on the collision zone between
Peninsular India and Asia in the Himalayan Orogen have proposed crustal shortening of c. 550 km (Ratsbacher et al. 1994). If the crustal shortening had begun at c. 590 Ma as suggested by the metamorphic ages in the Nampula Block, at a plate movement rate of c. 4 cm a21, similar to that recorded in current studies of the Himalayas (Ratsbacher et al. 1994), the emplacement of the
114
G. H. GRANTHAM ET AL.
Fig. 11. Schematic cross-section representing the geological relationships from northern Mozambique to southern Kirwanveggen, Antarctica. The cross-section summarizes the deformational structures, the distribution of the 495–530 Ma granitic intrusions in the footwall resulting from burial heating, and the setting for the isothermal decompression followed by isobaric cooling paths of the hanging-wall Namuno Block. The cross-section also shows the relative difference in erosion levels between Africa and Antarctica.
mega-nappe would have required c. 15 Ma to cover the possible 600 km envisaged. The nappe emplacement would then have been followed by inversion, isostatic uplift and erosion, with anatexis at depth c. 40 Ma after emplacement of the obducted slab, as suggested by England & Thompson (1986). The model of England & Thompson (1986) demonstrates that it would require c. 100 Ma for the thermal gradients of the footwall rocks to rise and achieve equilibrium with the hanging-wall rocks and that, depending on depth, it would take c. 40 Ma for rocks at middle crustal level of the footwall to heat into the field of anatexis where melting would be initiated. The model also demonstrates that, depending on the relative depth of melting, one can generate a wide range of granitoids depending on whether the melts are ‘minimum melts’ or are produced by vapour-absent dehydration melting resulting in charnockitic K-rich anhydrous melts. Although there appear to be different erosion levels in Africa and Antarctica, the implications of this model are that a substantial block of crust has been eroded and removed from this belt since c. 550 Ma, particularly in southern Africa. The recognition of reoriented fabrics in Sverdrupfjella and possibly northern Kirwanveggen suggests that these areas were probably in the footwall as well, providing an estimate of the area underlain by the nappe that is significantly larger than that reflected in Figure 4. The recognition of zircon populations with ages typical of the Namuno Block in the Transantarctic Mountains (Goodge 1997; Goodge et al. 2004) and similar rocks in Australia (Veevers et al. 2006) suggests that the sedimentary rocks now exposed in the Transantarctic Mountains and its extensions and the Ellsworth –Whitmore
Mountains (Flowerdew et al. 2007) were probably the depository of the erosion products from the collision of North and South Gondwana along the Damara –Zambezi–Lurio –Sri Lanka–central Dronning Maud axis. Another example of such deposition is in the Urfjell Group, in which the detrital zircon population (Fig. 12) has an age pattern unlike that of its surrounding Nampula Block type floor rocks (data from Croaker 1999) but a distribution more typical of the Namuno Block rocks from north of the Lurio Belt. The overthrust block of tectonic units belonging to the Namuno Block as envisaged in this paper would have placed the detrital source for the Urfjell Group significantly closer to the depository and, in view of the age of the Urfjell Group of c. 530 Ma, the location of the Urfjell Group provides an absolute southern limit of the extent of the overthrust block. This model also provides insights into the geochronological differences between the East African Orogen (Stern 1994; Meert 2003) and the Kungu Orogeny proposed and described by Meert (2003). Stern (1994) initially described the north – south-oriented East African Orogen based on fieldwork in North Africa, through Kenya and Tanzania and suggested that its timing was between c. 900 Ma and c. 550 Ma. The model presented here suggests that, except for the structural outliers in the Nampula Block, Antarctica and Sri Lanka, the East African Orogen is terminated along the Lurio Belt, where it is overprinted and reworked by the Kungu Orogen. The collisional front along the Damara –Zambezi–Lurio –Sri Lanka–central Dronning Maud axis is therefore equivalent to the Kungu Orogen described by Meert (2003). The rapid erosion and uplift of the Nampula Block is recorded in a titanite fission-track study by Daszinnies et al. (2006), who showed that the titanite fission-track ages north of the Lurio Belt
Fig. 12. Histogram and relative probability density curve of zircon ages from the Urfjell Group. Data from Croaker (1999).
ANTARCTIC–MOZAMBIQUE –SRI LANKA CORRELATION
are c. 300 Ma and become progressively younger to the coast, to c. 240 Ma. Erosion to surface was completed by c. 180 Ma, the age of the Karooage Angoche Andesite lavas on the northern Mozambique coast (Grantham et al. 2007b). This model explains many of the correlation conundrums that have puzzled scientists in Gondwana reconstructions. It needs to be tested with much additional work. This work should probably involve constraining the precise ages of deformation in the vicinity of the suture zones between the Namuno type blocks and Nampula type blocks in Mozambique, Malawi, Sri Lanka and Antarctica. The extent of the overthrust blocks should also be tested with detailed geochronology and mapping on the Urungwe klippe in northern Zimbabwe, as well as geochronology and P –T work on the nunataks at Mramornye in Dronning Maud Land. Another potential implication is that, after this collision, extensive parts of the northern Kalahari Craton may have been in the footwall of a large nappe. It is an open question as to whether there are other .550 Ma lithological units on the Zimbabwe Craton that may be allochthonous. Possible candidates include the Makuti Groups and the Rushinga Group in Zimbabwe and the Frontier Formation in Mozambique. The origin and distribution of the c. 850 Ma rocks in Mozambique adjacent to those in the vicinity of the NE corner of Zimbabwe need to be studied in detail. Geochronology focusing on low-temperature closure systems in Zimbabwe may provide some idea of how far south the nappe travelled over Zimbabwe, if at all. Similarly, the grits, sandstones, shales and conglomerates of the pre-Karoo-age Sijarira Group in northwestern Zimbabwe may represent the Cambrian-age detritus eroded from the overthrust Zambezi Belt rocks and, if so, then their current positions constrain the southern limit of the mega-nappe (P. Dirks pers. comm.;), just as the Urfjell Group do in southern Kirwanveggen, western DML. A sedimentological study combined with a zircon provenance study of the Sijarira Group would be particularly illuminating. The uncertainty of the position of the southwestern end of the Lurio Belt in central northern Mozambique and southern Malawi also needs further fieldwork and geochronology. We are sure that aspects of this model will be require revision and modification as new data are produced from further studies in Mozambique, Antarctica and Sri Lanka. New comprehensive geochronological and petrological data from Sri Lanka are vitally important. The model also needs to be tested with palaeomagnetic studies, particularly in view of the ‘destruction’, by crustal duplication followed by erosion, of large crustal blocks from the Namuno Block potentially involving 58 of latitude or 600 km.
115
We would like to acknowledge all the people who have over the years assisted in the field in various capacities, and who have assisted with data and sample collection in Antarctica and Mozambique. In Antarctica these include geologists R. Thomas and B. Groenewald, and in Mozambique these include R. Thomas, G. de Kock, M. du Toit, P. Botha, M. Kota, R. Opperman, M. Rohwer, J. C. Nolte, M. Cronwright, I. Haddon, J. Miller, S. De Azevedo, S. Fernando, R. Matola, G. Cune and S. Kagashima. We would also like to acknowledge the assistance of H. Kaidan and D. Dunkley with SHRIMP analyses at NIPR, Tokyo. Critical reviews by R. Hansen and S. Johnson significantly improved the manuscript. Discussions with P. Dirks provided valuable insights into aspects related to the Zimbabwe Craton cover. Correspondence with C. Reeves led to the recognition that the Lurio Belt shear zone could also have a horizontal attitude. Permission to publish the summarized geochronological data from Mozambique was also granted by E. Daudi, Director of Direca˜o Nacional Geologia, Maputo, Mozambique. This paper is dedicated to the pioneers R. Sacchi and P. Pinna, who probably would have developed this model had they had the access to extensive single zircon geochronology.
References A HRENDT , H., H UNZIKER , J. C. & W EBER , K. 1978. Age and degree of metamorphism and time of nappe emplacement along the southern margin of the Damara Orogen/Namibia. Geologische Rundschau, 67, 719– 742. A QUATER . 1983. Relatorio Final Vol. II—Cartografia Geologica. Cartografia Geologica e Prospecc¸ao Mineira e Geoquimica nas Provincias de Nampula e da Zambezia. A SAMI , M., S UZUKI , K. & G REW , E. S. 2005. Monazite and zircon dating by the chemical Th–U –total Pb isochron method (CHIME) from Aleysheyev Bight to the Sør-Rondane Mountains, East Antarctica: A reconnaissance study of the Mozambique suture in Eastern Queen Maud Land. Journal of Geology, 113, 59–82. B ABA , S., O WADA , M., G REW , E. & S HIRAISHI , K. 2006. Sapphirine– orthopyroxene–garnet granulite from Schirmacher Hills, Central Dronning Maud Land. In: F U¨ TTERER , D. K., D AMASKE , D., K LEINSCHMIDT , G., M ILLER , H. & T ESSENSOHN , F. (eds) Antarctica: Contributions to Global Earth Sciences. Springer, New York, 37–44. B ABA , S., O WADA , M. & S HIRAISHI , K. 2008. Contrasting metamorphic P– T path between Schirmacher Hills and Mu¨hlig-Hofmannfjella, Central Dronning Maud Land, East Antarctica. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: a Key to the East–West Gondwana Connection. Geological Society, London, Special Publications, 308, 401–418. B ARR , M. W. C. & B ROWN , M. A. 1987. Precambrian gabbro – anorthosite complexes, Tete Province, Mozambique. Geological Journal, 22, 139 – 159.
116
G. H. GRANTHAM ET AL.
B ARTON , C. M., E VANS , J. A., C ARNEY , J. N. & C ROW , M. J. 1993. Geological and structural framework of the Zambezi Belt, northeastern Zimbabwe. In: F INDLAY , R. H., U NRUG , R., B ANKS , M. R. & V EEVERS , J. J. (eds) Gondwana Eight. Balkema, Rotterdam, 55–68. B AUER , W., J ACOBS , J. & P AECH , H.-J. 2004. Structural evolution of the metamorphic basement of Central Dronning Maud Land, East Antarctica. Geologische Jahrbuch, B96, 325–363. B AUR , N., K RO¨ NER , A., L IEW , T. C., T ODT , W., W ILLIAMS , I. S. & H OFFMANN , A. W. 1991. U–Pb isotopic systematics of zircons from prograde and retrograde transition zones in high-grade orthogneisses, Sri Lanka. Journal of Geology, 99, 527–545. B INGEN , B. I., V IOLA , G., H ENDERSON , I. H. C. ET AL . 2006. Geochronology of Pan-African terrain assembly in NE Mozambique. In: Abstracts CAG 21, 3 –5 July, Maputo, Mozambique, 12–14. B ISNATH , A. & F RIMMEL , H. E. 2005. Metamorphic evolution of the Maud Belt: P– T– t path for highgrade gneisses in Gjelsvikfjella, Dronning Maud Land, East Antarctica. Precambrian Research, 43, 505– 524. B ISNATH , A., F RIMMEL , H. E., A RMSTRONG , R. A. & B OARD , W. S. 2006. Tectonothermal evolution of the Maud Belt: new SHRIMP U– Pb zircon data from Gjelsvikfjella, Dronning Maud Land, East Antarctica. Precambrian Research, 150, 95– 121. B JERKGARD , T., N ORDGULEN , O., B INGEN , B. I. ET AL . 2006. Overview of the Proterozoic geology in Niassa and Cabo Delgado Provinces, N. Mozambique. In: Abstracts CAG 21, 3–5 July, Maputo, Mozambique, 15–18. B OARD , W. S., F RIMMEL , H. E. & A RMSTRONG , R. A. 2005. Pan-African tectonism in the western Maud Belt: P –T – t path for high-grade gneisses in the H.U. Sverdrupfjella, East Antarctica. Journal of Petrology, 46, 671 –699. B ONIN , B. 2007. A-type granites and related rocks: Evolution of a concept, problems and prospects. Lithos, 97, 1–29. C ADOPPI , P., C OSTA , M. & S ACCHI , R. 1987. A cross section of the Namama Thrust belt (Mozambique). Journal of African Earth Sciences, 6, 493–504. C OSTA , M., C ADOPPI , P., S ACCHI , R. & F ANNING , C. M. 1994. U–Pb SHRIMP dating of zircons from Mozambique Gneiss. Bolletino della Societa` Geologica Italiana, 113, 173–178. C ROAKER , M. 1999. Geological constraints on the evolution of the Urfjell Group, southern Kirwanveggen, western Dronning Maud Land, Antarctica. MSc thesis, University of Natal, Pietermaritzburg. D ASZINNIES , M. C. I., E MMEL , B., J ACOBS , J. & G RANTHAM , G. 2006. Implications for Gondwana rifting in Northern Mozambique: constrains from titanite and apatite fission track thermochronology. In: Abstracts CAG 21, 3–5 July, Maputo, Mozambique, 35–36. D EWEY , J. F. 1988. Extensional collapse of orogens. Tectonics, 6, 1123–1139. D IRKS , P. H. G. M. & J ELSMA , H. A. 2006. The structural–metamorphic evolution of the northern margin of the Zimbabwe craton and the adjacent
Zambezi belt in northeastern Zimbabwe. In: Processes on the Early Earth. Geological Society of America, Special Papers, 405, 291– 313. D IRKS , P. H. G. M., J ELSMA , H. A., V INYU , M. L. & M UNYANYIWA , H. 1998. The structural history of the Zambezi Belt in northeast Zimbabwe; evidence for crustal extension during the early Pan-African. South African Journal of Geology, 101, 1 –16. D IRKS , P. H. G. M., K RO¨ NER , A., J ELSMA , H. A., S ITHOLE , T. A. & V INYU , M. L. 1999. Structural relations and Pb–Pb zircon ages for the Makuti Gneisses: evidence for a crustal-scale Pan African shear zone in the Zambezi Belt, northwest Zimbabwe. Journal of African Earth Sciences, 28, 427–442. E BY , G. N. 1992. Chemical subdivision of the A-type granitoids: Petrogenetic and tectonic implications. Geology, 20, 641– 644. E NGLAND , P. C. & T HOMPSON , A. 1986. Some thermal and tectonic models for crustal melting in continental collision zones. In: C OWARD , M. P. & R IES , A. C. (eds) Collision Tectonics. Geological Society, London, Special Publications, 19, 83–94. F LOWERDEW , M. J., M ILLAR , I. L., C URTIS , M. L., V AUGHAN , A. P. M., H ORSTWOOD , M. S. A., W HITEHOUSE , M. J. & F ANNING , C. M. 2007. Combined U– Pb geochronology and Hf isotope geochemistry of detrital zircons from early Paleozoic sedimentary rocks, Ellsworth –Whitmore Mountains block, Antarctica. Geological Society of America Bulletin, 119, 275– 288. G OODGE , J. 1997. Latest Neoproterozoic basin inversion of the Beardmore Group, central Transantarctic Mountains, Antarctica. Tectonics, 16, 682–701. G OODGE , J. W., W ILLIAMS , I. S. & M YROW , P. 2004. Provenance of Neoproterozoic and lower Paleozoic siliciclastic rocks of the central Ross orogen, Antarctica: Detrital record of rift-, passive-, and active-margin sedimentation. Geological Society of America Bulletin, 116, 1253–1279. G RANTHAM , G. H. 1992. Geological evolution of western H.U. Sverdrupfjella, Dronning Maud Land, Antarctica. PhD thesis, University of Natal, Pietermaritzburg. G RANTHAM , G. H., M OYES , A. B. & H UNTER , D. R. 1991. The age, petrogenesis and emplacement of the Dalmatian Granite, H. U. Sverdrupfjella, Dronning Maud Land, Antarctica. Antarctic Science, 3, 197–204. G RANTHAM , G. H., J ACKSON , C., M OYES , A. B., G ROENEWALD , P. B., H ARRIS , P. D., F ERRAR , G. & K RYNAUW , J. R. 1995. The tectonothermal evolution of the Kirwanveggan– H.U. Sverdrupfjella areas, Dronning Maud Land, Antarctica. Precambrian Research, 75, 209– 230. G RANTHAM , G. H., S TOREY , B. C., T HOMAS , R. J. & J ACOBS , J. 1997. The pre-breakup position of Haag Nunataks within Gondwana: possible correlatives in Natal and Dronning Maud Land. In: R ICCI , C. A. (ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antartica, Sienna, 13–20. G RANTHAM , G. H., E GLINGTON , B. M., T HOMAS , R. J. & M ENDONIDIS , P. 2001. The nature of the Grenvilleage charnockitic A-type magmatism from the Natal, Namaqua and Maud Belts of southern Africa and western Dronning Maud Land, Antarctica. Memoirs
ANTARCTIC–MOZAMBIQUE –SRI LANKA CORRELATION of National Institute of Polar Research, Special Issue, 55, 59– 86. G RANTHAM , G. H., M ABOKO , M. & E GLINGTON , B. M. 2003. A review of the evolution of the Mozambique Belt and implications for the amalgamation of Rodinia and Gondwana. In: Y OSHIDA , M., W INDLEY , B. F. & D ASGUPTA , S. (eds) Proterozoic East Gondwana: Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 401–426. G RANTHAM , G. H., A RMSTRONG , R. A. & M OYES , A. B. 2006. The age, chemistry and structure of mafic dykes at Roerkulten, H.U. Sverdrupfjella, western Dronning Maud Land, Antarctica. In: H ANSKI , E., M ERTANEN , S., R AMO , T. & V UOLLO , J. (eds) Dyke Swarms—Time Markers of Crustal Evolution. Proceedings of the Fifth/Fourth International Dyke Conference (IDC5), Rovaniemi, Finland. Balkema, Rotterdam, 213–224. G RANTHAM , G. H., I NGRAM , B. A., C RONWRIGHT , M. C. ET AL . 2007a. Map Explanation of Sheets Furancungo (1433) and Ulongue (1434). National Directorate of Geology, Republic of Mozambique, Maputo. G RANTHAM , G. H., M ACEY , P. H., I NGRAM , B. A. ET AL . 2007b. Map Explanation of Sheets Meconte (1439) and Nacala (1440). National Directorate of Geology, Republic of Mozambique, Maputo. G RAY , D. R., F OSTER , D. A., G OSCOMBE , B., P ASSCHIER , C. W. & T ROUW , R. A. J. 2006. 40 Ar/39Ar thermochronology of the Pan-African Damara Orogen, Namibia, with implications for tectonothermal and geodynamic evolution. Precambrian Research, 150, 49–72. G ROENEWALD , P. B., M OYES , A. B., G RANTHAM , G. H. & K RYNAUW , J. R. 1995. East Antarctic crustal evolution—geological constraints and modelling in western Dronning Maud Land. Precambrian Research, 75, 231–250. H ANSON , R. E., W ILSON , T. J. & M UNYANYIWA , H. W. 1994. Geologic evolution of the Neoproterozoic Zambezi orogenic belt in Zambia. Journal of African Earth Sciences, 18, 135–150. H ARGROVE , U. S., H ANSON , R. E., M ARTIN , M. W., B LENKINSOP , T. G., B OWRING , S. A., W ALKER , N. & M UNYANYIWA , H. 2003. Tectonic evolution of the Zambezi orogenic belt: geochronological, structural, and petrological constraints from northern Zimbabwe. Precambrian Research, 123, 159–186. H ARRIS , P. D. 1999. The geological evolution of Neumayerskarvet in the northern Kirwanveggen, western Dronning Maud Land, Antarctica. PhD thesis, Rand Afrikaans University, Johannesburg. H ARRIS , P. D., M OYES , A. B., F ANNING , C. M. & A RMSTRONG , R. A. 1995. Zircon ion microprobe results from the Maudheim High Grade Gneiss Terrane, Western Dronning Maud Land, Antarctica. In: Geocongress ’95 Abtracts, Rand Afrikaans University, Johannesburg, 240– 243. H IROI , Y., O GO , Y. & N AMBA , K. 1994. Evidence for prograde metamorphic evolution of Sri Lankan pelitic granulites, and implications for the development of continental crust. Precambrian Research, 66, 245–263. H OLLICK , L., B AUER , W., DE S OUSA S OARES , H. ET AL . 2006. Structural evolution in N. Mozambique—Milange, Mocuba and Insaca–Gurue Sheets. In: Abstracts CAG 21, 3– 5 July, Maputo, Mozambique, 68– 70.
117
H OLZL , S., H OFMANN , A. W., T ODT , W. & K OHLER , H. 1994. U –Pb geochronology of the Sri Lankan basement. Precambrian Research, 66, 123–149. J ACKSON , C. 1999. Characterisation of Mesoproterozoic to Paleoozoic crustal evolution of western Dronning Maud land. Unpublished report to South African National Antarctic Programme. Study 3. J ACKSON , C. & A RMSTRONG , R. A. 1997. The tectonic evolution of the central Kirwanveggen, Dronning Maud Land, Antarctica: Temporal resolution of deformation episodes using SHRIMP U– Pb zircon geochronology. In: Tectonics Division of the Geological Society of South Africa, 13th Anniversary Conference, Abstracts, 21– 22. J ACOBS , J., F ANNING , C. M., H ENJES -K UNST , F., O LESCH , M. & P AECH , H. J. 1998. Continuation of the Mozambique Belt into East Antarctica: Grenville age metamorphism and polyphase Pan-African high grade events in Central Dronning Maud Land. Journal of Geology, 106, 385 –406. J ACOBS , J., B AUER , W. & F ANNING , C. M. 2003a. Late Neoproterozoic/Early Paleozoic events in central Dronning Maud Land and significance for the southern extension of the East African Orogen into East Antarctica. Precambrian Research, 126, 27– 53. J ACOBS , J., B AUER , W. & F ANNING , C. M. 2003b. New age constraints for Grenvillian age metamorphism in western central Dronning Maud land (East Antarctica) and implications for the paleogeography of Kalahari in Rodinia. Geologische Rundschau, 92, 301– 315. J ACOBS , J., K LEMD , R., F ANNING , C. M., B AUER , W. & C OLOMBO , F. 2003c. Extensional collapse of the late Neoproterozoic–early Paleozoic East African– Antarctic Orogen in Central Dronning Maud Land, East Antarctica. In: Y OSHIDA , M., W INDLEY , B. F. & D ASGUPTA , S. (eds) Proterozoic East Gondwana: Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 271– 287. J AMAL , D. L. 2005. Crustal studies across selected geotransects in NE Mozambique: Differentiating between Mozambiquan (Kibaran) and Pan African events, with implications for Gondwana studies. PhD thesis, University of Cape Town. J OHNSON , S. P., R IVERS , T. & DE W AELE , B. 2005. A review of the Mesoproterozoic to early Palaeozoic magmatic and tectonothermal history of south–central Africa: implications for Rodinia and Gondwana. Journal of the Geological Society, London, 162, 433–450. J OHNSON , S. P., DE W AELE , B. & L IYUNGU , K. A. 2006. U– Pb sensitive high resolution ion microprobe (SHRIMP) zircon geochronology of granitoid rocks in eastern Zambia: Terrane subdivision of the Mesoproterozoic Southern Irumide Belt. Tectonics, 25, TC6004, doi:10.1029/2006TC001977. J OURDE , G. & V IALETTE , Y. 1980. La chaıˆne du Lurio (Nord Mozambique). Un te´moin de l’existence de chaıˆnes Kibariennes (800–1350 Ma) en Afrique Orientale. BRGM Internal Report. K LEINSCHRODT , R. 1994. Large-scale thrusting in the lower crustal basement of Sri Lanka. Precambrian Research, 66, 39–57.
118
G. H. GRANTHAM ET AL.
K OISTINEN , T. J., L EHTONEN , M. I., C UNE , G. ET AL . 2006. Geological mapping east and north of the Archaean Zimbabwe Craton, NW Mozambique. In: Abstracts CAG 21, 3–5 July, Maputo, Mozambique, 92. K RETZ , R. 1983. Symbols for rock-forming minerals. American Mineralogist, 68, 277–279. K RIEGSMAN , L. M. 1995. The Pan-African event in East Antarctica: a view from Sri Lanka and the Mozambique Belt. Precambrian Research, 75, 263– 277. K RO¨ NER , A. 1991. African linkage of Precambrian Sri Lanka. Geologische Rundschau, 80, 429–440. K RO¨ NER , A. 2001. Du Toit Memorial Lecture 1999: The Mozambique belt of East Africa and Madagascar: significance of zircon and Nd model ages for Rodinia and Gondwana supercontinent formation and dispersal. South African Journal of Geology, 104, 151– 166. K RO¨ NER , A., W ILLIAMS , I. S., C OMPSTON , W., B AUR , N., V ITANAGE , P. W. & P ERERA , L. R. K. 1987. Zircon ion microprobe dating of high grade rocks in Sri Lanka. Journal of Geology, 95, 775– 791. K RO¨ NER , A., J AECKEL , P. & W ILLIAMS , I. S. 1994. Pb-loss patterns in zircons from a high-grade metamorphic terrain as revealed by different dating methods: U–Pb and Pb–Pb for igneous and metamorphic zircons from northern Sri Lanka. Precambrian Research, 66, 151–181. K RO¨ NER , A., S ACCHI , R., J AECKEL , P. & C OSTA , M. 1997. Kibaran magmatism and Pan African granulite metamorphism in northern Mozambique: single zircon ages and regional implications. Journal of African Earth Sciences, 25, 467 –484. K RO¨ NER , A., W ILLNER , A. P., H EGNER , E., J AECKEL , P. & N EMCHIN , A. 2001. Single zircon ages, PT evolution and Nd isotopic systematics of high grade gneisses in southern Malawi and their bearing on the evolution of the Mozambique belt in southeastern Africa. Precambrian Research, 109, 257– 291. K RYNAUW , J. R. & J ACKSON , C. 1996. Geological evolution of western Dronning Maud Land within a Gondwana framework. South African National Antarctic Programme Final Report 1991– 1996. Geology Subsection, 1 –48. L AWVER , L. A., G AHAGEN , L. M. & D ALZIEL , I. W. D. 1998. A tight fit early Mesozoic Gondwana, a plate reconstruction perspective. Memoirs of National Institute of Polar Research, Special Issue, 53, 214– 229. M ACEY , P. H., I NGRAM , B. A., C RONWRIGHT , M. S. ET AL . 2007. Map Explanation of Sheets Alto Molo´cue` (1537), Murrupula (1538), Nampula (1539), Mogincual (1540), Errego (1637), Gile´ (1638) and Angoche (1639– 40). National Directorate of Geology, Republic of Mozambique, Maputo. M ANHICA , A. S. T. D., G RANTHAM , G. H., A RMSTRONG , R. A., G UISE , P. G. & K RUGER , F. J. 2001. Polyphase deformation and metamorphism at the Kalahari Craton—Mozambique Belt boundary. In: M ILLER , J. A., H OLDSWORTH , R. E., B UICK , I. S. & H AND , M. (eds) Continental Reactivation and Reworking. Geological Society, London, Special Publications, 184, 303–321. M A¨ NTTA¨ RRI , I., K OISTINEN , T., L EHTONEN , M. I., M ANNINEN , T., M AKITIE , H., H UHMA , H. & K UOSMANEN , E. 2006. U– Pb and Sm–Nd ages
for 27 magmatic rocks, NW Mozambique. In: Abstracts CAG 21, 3 –5 July, Maputo, Mozambique, 203–205. M ARTIN , H. 1974. Damara rocks as nappes on the Naukluft Mountains, South West Africa. Precambrian Research Unit, University of Cape Town, Bulletin, 15, 153–165. M EERT , J. 2003. A synopsis of events related to the assembly of eastern Gondwana. Tectonophysics, 362, 1– 40. M ELEZHIK , V. A., K UZNETSOV , A. B., F ALLICK , A. E., S MITH , R. A., G OROKHOV , I. M., J AMAL , D. & C ATUANE , F. 2006. Depositional environments and an apparent age for the Geci meta-limestones: Constraints on the geological history of northern Mozambique. Precambrian Research, 148, 19– 31. M OYES , A. B. & H ARRIS , P. D. 1996. Geological Evolution of Western Dronning Maud Land within a Gondwana Framework. South African National Antarctic Programme Final Report 1991–1996. Radiogenic Isotope Project, 1– 38. M OYES , A. B., K NOPER , M. W. & H ARRIS , P. D. 1997. The age and significance of the Urfjell Group, western Dronning Maud Land. In: R ICCI , C. A. (ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antartica, Sienna, 31– 36. P AULSSON , O. & A USTRHEIM , H. 2003. A geochronological and geochemical study of rocks from Gjelsvikfjella, Dronning Maud Land, Antarctica— implications for Mesoproterozoic correlations and assembly of Gondwana. Precambrian Research, 125, 113–138. P IAZOLO , S. 2004. Detailed geological mapping of the Mramornye nunataks and the western part of Schirmacheroase, Central Dronning Maud Land, East Antarctica. Geologische Jahrbuch, B96, 469–497. P INNA , P. 1995. On the dual nature of the Mozambique Belt, Mozambique to Kenya. Journal of African Earth Sciences, 21, 477– 480. P INNA , P. & M ARTEAU , P. 1987. Carta geologica de Moc¸ambique, 1 : 1 000 000 scale, with explanatory notes. Instituto Nac¸ional de Geologica, Maputo. P INNA , P., J OURDE , G., C ALVEZ , J. Y., M ROZ , J. P. & M ARQUES , J. M. 1993. The Mozambique Belt in northern Mozambique: Neoproterozoic (1100–850 Ma) crustal growth and tectogenesis, and superimposed Pan-African (800–550 Ma) tectonism. Precambrian Research, 62, 1–59. R AASE , P. & S CHENK , V. 1994. Petrology of granulitefacies metapelites of the Highland Complex, Sri Lanka: implications for the metamorphic zonation and the P– T path. Precambrian Research, 66, 265–294. R ATSBACHER , L., F RISCH , W., L IU , G. & C HEN , C. 1994. Distributed deformation in southern and western Tibet during and after the India–Asia collision. Journal of Geophysical Research, 99, 19917–19946. R AVIKANT , V., B HASKAR R AO , Y. J. & G OPALAN , K. 2002. P –T–t history of granulites from the Schirmacher Oasis, East Antarctica. 12th Annual V. M. Goldschmidt Conference Abstracts. Geochimica et Cosmochimica Acta, 66, 793. R AVIKANT , V., B HASKAR R AO , Y. J. & G OPALAN , K. 2004. Schirmacher Oasis as an extension of the
ANTARCTIC–MOZAMBIQUE –SRI LANKA CORRELATION Neoproterozoic East African Orogen into Antarctica: New Sm–Nd isochron age constraints. Journal of Geology, 112, 607– 616. R AVIKANT , V., L AUX , J. H. & P IMENTEL , M. 2008. Sm– Nd and U–Pb isotopic constraints for crustal evolution during Late Neoproterozoic from rocks of the Schirmacher Oasis, East Antarctica: geodynamic development coeval with the East African Orogeny. US Geological Survey, Open-File Report 2007-1047, Geological Survey and the National Academies, USES OF-2007-1047, Short Research Paper 007; doi:10.3133/of2007-1047.srp007. R EEVES , C. & DE W IT , M. 2000. Making ends meet in Gondwana: retracing the transforms of the Indian Ocean and reconnecting continental shear zones. Tectonics, 12, 272–280. R OBERTS , M. P., G RANTHAM , G. H., C RONWRIGHT , M. ET AL . 2005. Preliminary pressure temperature determinations on granulite-facies rocks of the Mugeba Klippe, north central Moc¸ambique. In: GEO2005 Extended Abstracts, Durban, South Africa, 193– 194. S ACCHI , R., M ARQUES , J., C OSTA , M. & C ASATI , C. 1984. Kibaran events in the southern Mozambique Belt. Precambrian Research, 25, 141 –159. S ACCHI , R., C ADOPPI , P. & C OSTA , M. 2000. Pan-African reactivation of the Lurio segment of the Kibaran Belt system: a reappraisal from recent age determinations in northern Mozambique. Journal of African Earth Sciences, 30, 629–639. S CHUMACHER , R., S CHENK , V., R AASE , P. & V ITANAGE , P. W. 1990. Granulite facies metamorphism of metabasic and intermediate rocks in the Highland series of Sri Lanka. In: A SHWORTH , J. R. & B ROWN , M. (eds) High grade Metamorphism and Crustal Anatexis. Unwin Hyman, Boston, MA, 235–271. S HACKLETON , R. M., V AIL , J. R. & W OOD , D. S. 1966. Preliminary report on the origin and significance of the Urungwe Klippe, Rhodesia. Annual Report, Research Institute of African Geology, University of Leeds, 10, 10– 13. S HIRAISHI , K. & K AGAMI , H. 1992. Sm– Nd and Rb–Sr ages of metamorphic rocks from the Sør Rondane Mountains, east Antarctica. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (ed.) Recent Progress in Antarctic Earth Science. Terra, Tokyo, 29– 35. S HIRAISHI , K., A SAMI , M., I SHIZUKA , H. ET AL . 1991. Geology and metamorphism of the Sør Rondane mountains, east Antarctica. In: T HOMSON , M. R. A., C RAME , J. A. & T HOMSON , J. W. (eds) Geological Evolution of Antarctica. Cambridge University Press, Cambridge, 77–82. S HIRAISHI , K., D UNKLEY , D., H OKADA , T., F ANNING , C. M., K AGAMI , H. & H AMAMOTO , T. 2008. Geochronological constraints of the Late Proterozoic–Cambrian
119
crustal evolution in eastern Dronning Maud Land, East Antarctica: a synthesis of SHRIMP U– Pb age and Nd model age data. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: a Key to the East–West Gondwana Connection. Geological Society, London, Special Publications, 308, 21– 68. S IEGFRIED , P. R. 1999. The Monapo structure and intrusive complex—An example of large scale alkaline metasomatism in northern Mozambique. In: S TANLEY , C. J. ET AL . (eds) Mineral Deposits: Processes to Processing. Balkema, Rotterdam, 683– 686. S IRCOMBE , K. 2000. The usefulness and limitations of binned frequency histograms and probability density distributions for displaying absolute age data. Radiogenic age and isotopic studies. Geological Survey of Canada, Report, 13. S TERN , R. J. 1994. Arc assembly and continental collison in the Neoproterozoic East African Orogen: Implications for the consolidation of Gondwanaland. Annual Review of Earth Science, 22, 319– 351. T HOMAS , R. J., B AUER , W., B INGEN , B. ET AL . 2006. Mozambique Belt in the Milange–Mocuba– Malemba area, Mozambique. In: Abstracts CAG 21, 3– 5 July, Maputo, Mozambique, 161–163. V EEVERS , J. J., B ELOUSOVA , E. A., S AEED , A., S IRCOMBE , K., C OOPER , A. F. & R EAD , S. E. 2006. Pan-Gondwanaland detrital zircons from Australia analysed for Hf-isotopes and trace elements reflect an ice-covered Antarctic provenance of 700–500 Ma age, TDM of 2.0– 1.0 Ga, and alkaline affinity. Earth-Science Reviews, 76, 135– 174. V INYU , M. L., H ANSEN , R. E., M ARTIN , M. W., B OWRING , S. A., J ELSMA , H. A., K ROL , M. A. & D IRKS , P. H. G. M. 1999. U– Pb and 40Ar/39Ar geochronological constraints on the tectonic evolution of the eastern part of the Zambezi orogenic belt, northeast Zimbabwe. Precambrian Research, 98, 67– 82. V IOLA , G., H ENDERSON , I., B INGEN , B., F EITIO , P., T HOMAS , R. J., H OLLICK , L. & J ACOBS , J. 2006. A new tectonic framework for northern Mozambique. In: Abstracts CAG 21, 3 –5 July, Maputo, Mozambique, 168. W AREHAM , C. D., P ANKHURST , R. J., T HOMAS , R. J., S TOREY , B. C., G RANTHAM , G. H., J ACOBS , J. & E GLINGTON , B. M. 1998. Pb, Nd, and Sr isotope mapping of Grenville-age crustal provinces in Rodinia. Journal of Geology, 106, 647–659. W ESTERHOF , A. B. 2006. Tectonic framework of central and southern Mozambique in relation to Africa’s major building blocks. In: Abstracts CAG 21, 3– 5 July, Maputo, Mozambique, 174– 177. W ILSON , T. J., G RUNOW , A. M. & H ANSON , R. E. 1997. Gondwana assembly: the view from southern Africa and East Gondwana. Journal of Geodynamics, 23, 263–268.
An overview of geological studies of JARE in the Napier Complex, Enderby Land, East Antarctica HIDEO ISHIZUKA Department of Geology, Kochi University, Kochi 780-8520, Japan (e-mail:
[email protected]) Abstract: Subsequent to the reconnaissance fieldwork in 1982, the Japanese Antarctic Research Expedition (JARE) carried out extensive geological studies that focused on structural and tectonic aspects, petrology, geochemistry and geochronology of the Napier Complex in Enderby Land, East Antarctica. Detailed field investigations in several key areas, including geological mapping of the Mt. Riiser-Larsen area and Tonagh Island, revealed that the Napier Complex comprises layered and massive gneiss units, of which the layered unit is composed of garnet felsic gneiss, orthopyroxene felsic gneiss, pelitic and basic gneisses, impure quartzite, and minor metamorphosed banded iron formation, whereas the massive unit consists mainly of orthopyroxene felsic gneiss. The boundary between the units is transitional in the Mt. Riiser-Larsen area, in which metamorphosed anorthosite and ultramafic rocks occur as thin layers, or blocks or pods, but on Tonagh Island the boundary is closely associated with the shear zone. Nine deformation episodes (D1 –D9) were suggested for Tonagh Island. These results of fieldwork were presented in detail in two geological maps. Geochemical studies showed that (1) garnet–sillimanite gneisses and garnet-rich felsic gneisses were derived from mudstone and sandstone, respectively, both enriched in MgO, Cr and Ni; (2) orthopyroxene felsic gneisses have a close REE affinity with Archaean tonalite–trondhjemite–granodiorite (TTG); (3) basic gneisses were derived from light rare earth element (LREE)-enriched or -depleted basalts; (4) meta-ultramafic rocks are comparable with komatiite and related depleted mantle peridotite. This suite of protoliths is reminiscent of Archaean greenstone–granite belts. Precise analyses of physical conditions of metamorphism were carried out by using reliable approaches such as feldspar thermometry, alumina content of orthopyroxene, inverted pigeonite and bulk-rock compositions, and clinoand orthopyroxene compositions with different textures (porphyroblastic and neoblastic), and the results suggested that the maximum metamorphic temperature might have reached 1130 8C (i.e. ultrahigh-temperature (UHT) metamorphism). P– T evolution of the Napier UHT metamorphism was examined by analyses of reaction textures combined with fluid inclusion studies, suggesting both clockwise (Bunt Island) and counterclockwise (Mt. Riiser-Larsen and Tonagh Island) P– T– t paths. U–Pb sensitive high-resolution ion microprobe and secondary ionization mass spectrometry zircon ages from the Mt. Riiser-Larsen area and Tonagh Island indicate three stages of protolith formation at around 3.28 –3.23, 3.07 and 2.68–2.63 Ga, and two contrasting ages for the timing of peak UHT metamorphism at either c. 2.55 or c. 2.51– 2.45 Ga. On the basis of these results, more comprehensive studies on the Napier Complex are essential in the future for understanding (1) the role and age of TTG protolith and (2) the origin and timing of UHT metamorphism.
In Enderby Land, East Antarctica, the Napier Complex is situated in the northern part and the Rayner Complex in the southern part (Fig. 1). After a reconnaissance investigation by Soviet pioneers (Kamenev 1975; Ravich & Kamenev 1975), regular fieldwork on the Napier Complex was carried out by Australian expeditions during 1970–1980. The results have been summarized by Grew & Manton (1979), Sheraton et al. (1980, 1987), Grew (1982a) and Harley & Hensen (1990). These studies have revealed that (1) the Napier Complex underwent extremely hightemperature metamorphism (.900 8C), classified as ultrahigh-temperature (UHT) metamorphism (e.g. Spear 1993), during the late Archaean era,
and (2) a part of the protoliths of the Napier Complex shows a very old radiometric age, going back to early Archaean time, and thus belongs to one of the oldest continental crusts in the world. It is, therefore, most likely that the Napier Complex offers an excellent opportunity to study the processes of continental birth and subsequent growth during the Archaean to Proterozoic era. With the aim of further understanding such processes, the 23rd Japanese Antarctic Research Expedition (JARE-23) started geological investigations in the Napier Complex in 1982, and this work was followed by several expeditions in later years (Table 1). In particular, during 1996–1999 full summer field studies were carried out in the
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 121 –138. DOI: 10.1144/SP308.5 0305-8719/08/$15.00 # The Geological Society of London 2008.
122
H. ISHIZUKA
Fig. 1. Locality map of the Napier and Rayner Complexes, Enderby Land, East Antarctica.
Napier Complex (JARE-38, 39, and 40) under the auspices of a multidisciplinary project of the National Institute of Polar Research (NIPR), Japan, to study the Structure and Evolution of East Antarctic Lithosphere (‘SEAL’ project). In this study, an overview of important results of JARE studies on the Napier Complex obtained mainly as part of the SEAL project is provided, along with a brief consideration of future studies.
Lithologies and geological structures On the basis of field studies in the Napier Complex supported by JARE-38 (Ishizuka et al. 1997a, b), -39 (Moriwaki 1998) and -40 (Motoyoshi et al. 1999), detailed geological maps of the Mt. Riiser-Larsen
area (Ishizuka et al. 1998; Ishikawa et al. 2000) and Tonagh Island (Osanai et al. 1999, 2001a) were published (Fig. 2). These geological maps are accompanied by explanatory text that describes the geological structure and lithology of the area, and includes preliminary data on petrology, geochemistry and geochronology, which are briefly outlined below. The Mt. Riiser-Larsen area is underlain by UHT metamorphic rocks and minor dolerite dykes; the dykes are provisionally termed ‘Amundsen dykes’ after Sheraton et al. (1987), and were apparently emplaced after the UHT metamorphism (Ishizuka et al. 1998; Ishikawa et al. 2000). Suzuki et al. (2008) studied the Amundsen dykes in this area in detail, and revealed that: (1) there are NE–SWand north– south-striking dykes, of which the
Table 1. Localities where JARE parties have surveyed, and localities where icebreaker Shirase helicopters reconnaissance flights have landed in the Napier and Rayner Complexes JARE
Date
23 29 31
1982.2 1988.2 1990.2
34
1993.2
35 36 37 38 39
1994.2 1995.2 1996.2 1996.12 –1997.2 1998.1 –1998.2
40
1998.12 –1999.1
41 42 46
2000.2 2000.12 –2001.2 2005.2
Research areas Mt. Riiser-Larsen Mt. Riiser-Larsen Mt. Riiser-Larsen, Edward Is., Mt. Oldfield, Tonagh Is., Mt. Pardoe, Hydrographer Is., McIntyre Is. Hydrographer Is., McIntyre Is., Mt. Riiser-Larsen, Forefinger Point, Raggatt Mts, Dick Peaks,* Mt. Humble,* Mt. Maslen* Mt. Riiser-Larsen Mt. Riiser-Larsen Mt. Riiser-Larsen Mt. Riiser-Larsen, Tonagh Is. Tonagh Is., Bunt Is., Priestley Peak, Beaver Is., Bowl Is., Mt. Trail, Mt. Tod, Mt. Riiser-Larsen Mt. Riiser-Larsen, Tonagh Is., Mt. Pardoe, Howard Hills, Edward Is., Christmas Point Tonagh Is. Mt. Riiser-Larsen, Tonagh Is. Condon Hills, Mt. Lira, Mt. Yuzhnaya, Mt. Bergin, Forefinger Point, Fyfe Hills, Mt. Cronus
*Areas where icebreaker Shirase helicopters landed.
JARE STUDIES IN THE NAPIER COMPLEX
123
Fig. 2. Locality map of the northwestern part of the Napier Complex.
north–south-striking dykes interrupt the NE –SWstriking ones; (2) these dykes are geochemically grouped into tholeiite basalt (THB), highmagnesian andesite (HMA), alkaline basalt (AL), and enriched mid-ocean ridge basalt (E-MORB)like rock (THB-m), of which the THB and HMA belong to the NE –SW-striking dykes, and the AL and THA-m to the north–south-striking varieties; (3) the Sm–Nd and Rb –Sr bulk-rock isotope ratios of the THB dykes define an isochron ages of 2.0–1.9 Ga, whereas the AL dykes yield an isochron age of 1.2 Ga in the Rb –Sr system. Significantly, the AL and THB-m dykes are very rare in other areas of the Napier Complex (Sheraton et al. 1987), and the emplacement age of 2.0–1.9 Ga is the first report for the Amundsen dykes from the Napier Complex (Sheraton & Black 1981; Sheraton et al. 1987). Somewhat surprisingly, however, the 2.0–1.9 Ga period coincides with a global peak in mantle-derived magmatism (Condie 1997). Suzuki et al. (2008) consequently suggested that: (1) the north–south-striking dykes may occur in restricted areas in the Napier Complex, whereas the
NE–SW-striking dykes are regional; (2) the 2.0–1.9 Ga magmatism of the NE–SW-striking dykes may have been related to the formation of continental crust of the Rayner Complex. Furthermore, on the basis of geochemical affinities of the THB and HMA dykes, such as large ion lithophile element (LILE) and light rare earth element (LREE) enrichment and negative anomalies of Nb, Ti and/or P in a spider diagram normalized to primitive mantle, Suzuki et al. (2008) demonstrated that the origin of these dykes is analoguous to modern subduction-related arc volcanism. The UHT metamorphic rocks in the Mt. RiiserLarsen area commonly display metamorphic foliation that strikes NE –SW to east– west, and dips at a moderate to gentle angle (20 –408) to the south or SE, and they can be divided into layered and massive gneiss units (Ishizuka et al. 1998; Ishikawa et al. 2000). The layered gneiss unit, occurring in the central to northwestern part of the area, is characterized by layering composed of garnet felsic gneiss with subordinate amounts of orthopyroxene felsic gneiss, pelitic and basic
124
H. ISHIZUKA
gneisses, impure quartzite, and metamorphosed banded iron formation. The massive gneiss unit, developing in the southern to southeastern part of the area, consists mainly of orthopyroxene felsic gneiss, in which the layering is not conspicuous and the lithology is rather monotonous. As noted by Sheraton et al. (1987), this essentially follows the Soviet geologists’ subdivision of the Napier Complex into two major gneiss groups: layered garnet –quartz –feldspar gneiss with subordinate pelitic, psammitic, and ferruginous metasediments, and massive pyroxene –quartz –feldspar gneiss with minor mafic granulite (Kamenev 1975). However, these two gneiss units are not mutually exclusive. Indeed, transitional varieties occur between them, in which metamorphosed anorthosite and metaultramafic rocks (clinopyroxenite, orthopyroxenite and peridotite) occur characteristically as thin layers, or blocks or pods. The most prominent geological structure in the Mt. Riiser-Larsen area could be denoted by shear zones, in which the metamorphic rocks have been sheared to mylonites or sometimes pseudotachylites. These shear zones are mostly near-vertical and tend to follow pre-existing structures such as dyke margins. The width of the shear zones ranges from several centimetres to a few metres, but in the western part of the area a shear zone with the maximum width of 400 m occurs, striking north to south; Ishizuka et al. (1998) and Ishikawa et al. (2000) called this shear zone the Riiser-Larsen Main Shear Zone (RLMSZ). The lithology and geological structure are discontinuous from the western to the eastern parts of the RLMSZ. Ishikawa et al. (2000) found that two phases of folding are present in both parts of RLMSZ: the first is a fold with the axes aligned NNE–SSW and wavelengths up to 100 m, and the second is a fold that forms a broad dome structure after the UHT conditions. On Tonagh Island the metamorphic rocks are divided into five lithological units (Units I –V from north to south) based on their lithologies and geological structures (Osanai et al. 1999, 2001a). Thrust-shear zones accompanied by remarkable mylonite and later pseudotachylite–cataclasite bound each unit. Of these units, Unit I consists mainly of layered gneisses similar to the layered gneiss unit of the Mt. Riiser-Larsen area, whereas Units II, III and IV comprise two-pyroxene gneiss and garnet –orthopyroxene gneiss with minor layered gneisses. Unit V is mainly composed of orthopyroxene- and garnet-bearing quartzofeldspathic gneisses with subordinate layered gneisses. This lithological contrast between Unit V and other units led Osanai et al. (1999, 2001a) to suggest that the most prominent tectonic boundary on Tonagh Island may be the shear zone between Units V and IV. Subordinate amounts of
unmetamorphosed alkali-dolerite and granitic pegmatite cut across the sequence of metamorphic rocks, and the alkali-dolerite is also discordant to the unit boundary shear zone. Toyoshima et al. (1999) analysed the geological structure of Tonagh Island, and divided the deformation history into D1 –D9; the D1 structure would have been formed under non- or weakly deformational conditions during the thermal peak of prograde metamorphism, the D2 –D6 structures would have been produced under retrograde granulite-facies conditions, and subsequently the D7 –D9 brittle faulting modified the structures in part. For the whole Napier Complex, Sheraton et al. (1987) previously proposed three tectonothermal episodes; the first episode (D1 –M1), characterized by the formation of the present foliation, intrafolial folds and lineation, was synchronous with the peak of the granulite-facies metamorphism; the second episode (D2 –M2) produced tight to isoclinal, commonly asymmetric folds under granulite-facies conditions; and the third episode (D3 –M3) formed major asymmetric folding during the waning stage of high-grade metamorphism. Toyoshima et al. (1999) interpreted their D1, D2 –D4, and D5 – D6 stages as corresponding to the first (D1 –M1), second (D2 –M2) and third (D3 –M3) episodes, respectively. Toyoshima et al. (2001) further suggested multiple stages of pseudotachylite formation, related to D3, D6 and D8 deformations. Thus, detailed geological information for two key areas in the Napier Complex, the Mt. Riiser-Larsen area and Tonagh Island, were obtained during the JARE field studies, which will aid our understanding of the processes of Archaean crustal formation and UHT metamorphism. However, the regional geological map of Sheraton et al. (1987) is the only available regional geotectonic information on the Napier Complex, and there is still a lack of detailed geological information for other regions in Enderby Land. In this regard, the structural data for the Napier Complex as shown by Toyoshima et al. (2008, Figs 1 and 5) show that: (1) the Napier Complex can be subdivided into several units or blocks separated by east –west, NE –SW-, and NW– SE-striking faults including the RLMSZ; (2) the general strike changes from east –west in the western part to NNE– SSW or NE–SE in the central to eastern part; (3) several folds trending east –west and NE– SW are developed; (4) a dome-and-basin fold pattern on a regional scale is characteristic in some areas; (5) these structural subdivisions may be closely related to P–T –t evolution such as a clockwise or counterclockwise P–T –t path. These structural features are constructed from attitude data for foliations shown in the geological maps of Sheraton et al. (1987), Ishikawa et al. (2000),
JARE STUDIES IN THE NAPIER COMPLEX
and Osanai et al. (2001a), which will be useful for deducing a regional tectonic framework of the Napier Complex in the future.
Protoliths and their formation ages Geochemical studies on protoliths of the Napier UHT metamorphic rocks in the Mt. Riiser-Larsen area by Suzuki et al. (1999) revealed that: (1) garnet –sillimanite gneisses and garnet-rich felsic gneisses are of sedimentary origin, such as mudstone and sandstone, respectively, which are characteristically enriched in MgO (c. 4.1 wt%), Cr (c. 680 ppm) and Ni (c. 130 ppm) in the garnet –sillimanite gneisses, and in MgO (c. 2.7 wt%), Cr (c. 170 ppm) and Ni (c. 70 ppm) in the garnet-rich felsic gneisses; (2) orthopyroxene felsic gneisses and garnet-poor felsic gneisses are chemically comparable with CIPW normative tonalite to granodiorite and granite, respectively, and have a close REE affinity with Archaean tonalite –trondhjemite– granodiorite (TTG) (Luais & Hawkesworth 1994); (3) mafic gneisses are divided into a quartz-free and LREE-depleted type and a quartz-bearing and LREE-enriched type, suggesting a different source material for these two types; (4) phlogopite-free and phlogopitebearing meta-ultramafic rocks were derived from depleted mantle peridotites and komatiitic rocks, respectively, and the latter exhibit a magmatic differentiation controlled by olivine fractionation. The sedimentary precursors of these rocks (mudstone and sandstone enriched in MgO, Cr and Ni) are very similar to the sedimentary rocks reported from Archaean terranes (Condie 1997) such as the Kaapvaal Craton in southern Africa (Condie & Wronkiewicz 1990), where enrichment in these components appears to reflect the existence of komatiite –high-Mg basalt sources. For Tonagh Island, Owada et al. (1999, 2000) also studied the geochemistry of mafic gneisses and meta-ultramafic rocks (pyroxenite, websteritic peridotite and hornblende-bearing lherzolitic peridotite), and showed that (1) a majority of mafic gneisses are tholeiitic basalts in composition, and (2) some of the mafic gneisses and meta-ultramafic rocks are enriched in MgO (up to 31 wt%) and LREE, and resemble komatiitic basalts to komatiites. Of particular interest in the Riiser-Larsen area and on Tonagh Island is the presence of the TTG –komatiite association that is commonly reported from other Archaean greenstone –granite belts (e.g. Condie 1994). Sheraton et al. (1987) previously found that the less-fractionated high-Mg metamorphosed dykes are characterized by high MgO, Cr and Ni, but rather low TiO2, Na2O, P2O5, Zr, Nb and Y, and suggested that in these
125
respects these dykes have some chemical affinities with basaltic komatiites. Also, Sheraton et al. (1987) described the precursors of felsic gneisses throughout the Napier Complex, and suggested that they are comparable with the TTG protolith. It follows that, although the rocks were metamorphosed under UHT conditions, the protolith of the Napier Complex has an important role in understanding the continental evolution. On the other hand, the two types (LREE-enriched and -depleted types) of mafic gneisses described by Suzuki et al. (1999) and Owada et al. (1999, 2000) were reported by Sheraton et al. (1987) from other areas of the Napier Complex. This means that there are two types of source materials for the mafic gneisses throughout the Napier Complex, which is also important for understanding the origin and evolution of the Napier continental crust. Ages of protoliths were determined by means of the Sm–Nd bulk-rock method, U –Pb sensitive high-resolution ion microprobe (SHRIMP) and secondary ionization mass spectrometry (SIMS) mineral isotope methods, and the chemical U –Th –Pb isochron method (CHIME) (Table 2). The results show four age clusters at c. 3.7, 3.3, 3.0 and 2.6 Ga. Of these, the U –Pb SHRIMP and SIMS zircon ages are restricted to 3.28 –3.23 Ga (Shiraishi et al. 1997; Hokada et al. 2003), 3.07 Ga (Hokada et al. 2003) and 2.68 –2.63 Ga (Carson et al. 2002; Crowe et al. 2002). These U –Pb data are usually obtained from zircons with oscillatory zoning pattern and/or magmatic Th/U ratio, and are interpreted as the ages of protoliths. In the literature, apart from JARE studies, there are many geochronological studies focusing on the ages of protoliths in the Napier Complex. After reports of Pb –Pb ages of c. 4.0 Ga for orthogneisses from Fyfe Hills (Ravich & Kamenev 1975; Sobotovich et al. 1976), the subsequent studies by using U– Pb SHRIMP zircon dating indicated tonalitic igneous events at 4.0–3.8 Ga from Fyfe Hills, Mt. Sones and Gage Ridge (Black et al. 1986a; Harley & Black 1997; Kelly & Harley 2005). These very old ages have not been detected during JARE studies, although Owada et al. (1994) and Asami et al. (1998) reported an Sm–Nd bulk-rock age of 3.71 Ga for mafic gneiss from Tonagh Island and a CHIME zircon age of 3.65 Ga for felsic gneiss from Mt. Cronus, respectively. In contrast the U –Pb SHRIMP zircon ages of 3.28 –3.23 Ga from Tonagh Island (Shiraishi et al. 1997) and 3.27 Ga from Mt. Riiser-Larsen (Hokada et al. 2003) have not been found in other areas, although other radiometric data such as conventional zircon data, Rb–Sr bulk-rock data and Sm–Nd bulk-rock data showed the ages of protoliths to be 3.2 –3.0 Ga from Proclamation Island (Black et al. 1986b; Sheraton & Black
126
H. ISHIZUKA
Table 2. Age data estimated as protolith ages of the Napier Complex measured by the JARE geology group Sample locality
Method
Lithology
Age (Ga)
Tonagh Island
Sm –Nd bulk-rock
Tonagh Island
U –Pb (SHRIMP) mineral (core of zircon) U –Th–Pb (CHIME) mineral (core of zircon) Sm –Nd bulk-rock
Mafic gneiss Felsic gneiss Felsic gneiss
3.71 2.46 3.28– 3.23
Shiraishi et al. (1997)
Felsic gneiss
3.65
Asami et al. (1998)
Mafic gneiss Felsic gneiss Mafic gneiss Felsic gneiss
2.92 3.02 2.63 2.63
Suzuki (2000) Carson et al. (2002)
Felsic mylonite
2.68
Crowe et al. (2002)
Felsic gneiss
3.27
Hokada et al. (2003)
Felsic gneiss
3.07
Mt. Cronus Mt. Riiser-Larsen Tonagh Island Tonagh Island Mt. Riiser-Larsen
U –Pb (SIMS) mineral (core of zircon*) U –Pb (SHRIMP) mineral (core –margin of zircon*) U –Pb (SHRIMP) mineral (core of zircon*)
Reference Owada et al. (1994)
*Zircon with oscillatory zoning.
1983) and Fyfe Hills (Black et al. 1983, 1984). U– Pb SHRIMP zircon ages of 2.98–2.92 Ga, similar to that (3.07 Ga) from Mt. Riiser-Larsen (Hokada et al. 2003), have been also reported from Proclamation Island and Dallwitz Nunatak (Harley & Black 1997; Kelly & Harley 2005). As suggested by Hokada et al. (2003), there are at least three magmatic events in the Napier Complex, at c. 3.8 Ga in Fyfe Hills, Mt. Sones and Gage Ridge, 3.3 Ga in Tonagh Island and Mt. Riiser-Larsen and 3.0 Ga in Proclamation Island, Dallwitz Nunatak and Mt. Riiser-Larsen. In addition, it is most likely that the U–Pb SHRIMP zircon ages of about 2.63 and 2.68 Ga of felsic gneisses reported from Tonagh Island (Carson et al. 2002; Crowe et al. 2002) are the youngest ages of protoliths in the Napier Complex. It is, therefore, stressed that the JARE studies add new evidence of U–Pb SHRIMP zircon ages (3.3 and 2.6 Ga) to the field of magmatic activities of the Napier Complex. Of these, the 3.3 Ga event has been, however, reported as the age of protoliths from the Archaean terranes in eastern Australia (Kemp et al. 2006) and South Africa (Shirey et al. 2003).
UHT metamorphism and its peak age After the formation of protoliths, the Napier Complex underwent UHT metamorphism characterized by the presence of such diagnostic minerals and mineral assemblages as sapphirine þ quartz (Dallwitz 1968; Ellis 1980; Grew 1980; Harley & Hensen 1990), osumilite (Ellis 1980; Grew 1982b), inverted pigeonite (Sandiford & Powell 1986a; Harley 1987), and sillimanite þ orthopyroxene þ quartz (Harley 1985; Sheraton
et al. 1987). On the basis of mineral assemblages and conventional geothermobarometry, the highest metamorphic temperature (950 –1020 8C) was restricted to the northwestern area of the Napier Complex, that is, the area around Amundsen Bay, whereas the pressure conditions increased from the northern area (c. 5 kbar) to the southern area (11 kbar) of the Napier Complex (e.g. Harley & Hensen 1990; Harley 1998). Also, it has been proposed that the peak conditions of the Napier UHT metamorphism were followed by a period of near-isobaric cooling, for which the pressure was low in the north (,8 kbar) and high in the south (9–10 kbar) (Ellis 1980; Harley & Hensen 1990; Harley 1998). This gradation of pressures is suggested by regional differences in core Al2O3 contents of orthopyroxene equilibrated with garnet þ sillimanite (Harley 1985), differences in reaction coronas on sapphirine þ quartz such as cordierite þ sillimanite þ garnet in the north and orthopyroxene þ sillimanite þ garnet in the south (Sheraton et al. 1987), and the stability of osumilite þ garnet in the area to the north of Amundsen Bay compared with orthopyroxene þ sillimanite þ K-feldspar þ quartz southwards and eastwards (Hensen & Motoyoshi 1992; Harley 1998). These characteristics of UHT metamorphism in the Napier Complex were also reported by the JARE reconnaissance fieldwork in the Mt. Riiser-Larsen area (Motoyoshi & Matsueda 1984, 1987; Makimoto et al. 1989; Motoyoshi & Hensen 1989; Motoyoshi et al. 1990), and reviewed by Motoyoshi (1998). To clarify the more detailed metamorphic characteristics of the Napier Complex, the mineral assemblages, mineral chemistries, and geological structure in the Mt. Riiser-Larsen area (Ishizuka
JARE STUDIES IN THE NAPIER COMPLEX
et al. 1998; Hokada 1999; Ishikawa et al. 2000), Tonagh Island (Hokada et al. 1999; Osanai et al. 1999, 2001a; Owada et al. 1999; Toyoshima et al. 1999; Tsunogae et al. 1999), Howard Hills (Yoshimura et al. 2000; Miyamoto et al. 2004), and Christmas Point (Yoshimura et al. 2001) were studied in detail. On the basis of these descriptive studies, especially at the Mt. Riiser-Larsen area and Tonagh Island, several new approaches were applied for the determination of metamorphic temperature (Hokada 1999, 2001; Ishizuka et al. 1999, 2002; Harley & Motoyoshi 2000; Hokada & Suzuki 2006). Hokada (1999, 2001) calculated chemical compositions of reintegrated perthitic, mesoperthitic and antiperthitic feldspar minerals from felsic gneiss and garnet – orthopyroxene– sapphirine gneiss by the modal proportions and the chemical analyses of host and lamellar domains formed through exsolution, and estimated minimum equilibrium temperatures of 1000 – 1100 8C by using ternary feldspar solvus models. Similarly, Hokada & Suzuki (2006) recovered pre-exsolution single-phase compositions separately for the feldspar core and whole feldspar grain from felsic gneisses with tonalitic and granodioritic compositions, and calculated a temperature range of 940 –1100 8C for the feldspar core and 850 –1070 8C for the whole feldspar grain. Harley & Motoyoshi (2000) used the alumina content of porphyroblastic orthopyroxene (12.2 + 0.5wt% Al2O3) from the sapphirine þ orthopyroxene þ quartz granulite and calculations in the MAS and FMAS systems with theoretical considerations, and deduced a temperature in excess of 1120 8C. Ishizuka et al. (1999) analysed compositions of inverted pigeonite (exsolution intergrowth of clino- and orthopyroxenes) and bulk rocks from quartz –magnetite rocks, and, using ternary pyroxene phase diagrams, estimated a temperature of 1130 8C, at which one phase pyroxene (pigeonite) was once stable. Ishizuka et al. (2002) determined clino- and orthopyroxene compositions with different textural features (porphyroblastic and neoblastic) from meta-ultramafic rock, and used pyroxene thermometry to calculate a temperature that ranged up to 1130 8C for the porphyroblastic pyroxenes. The results of these new approaches revealed that the maximum temperature of the Napier UHT metamorphism reached 1130 8C at the Riiser-Larsen area and Tonagh Island. To further characterize the Napier UHT metamorphism, Hensen & Osanai (1994) and Motoyoshi & Hensen (2001) have carried out analyses of fluorine in mica minerals and related experiments. Hensen & Osanai (1994) found that synthetic F-rich phlogopite (Mg/(Fe þ Mg) ¼ 0.75) with F/(F þ OH) ¼ 0.6 (equivalent to 5–6 wt% F) disappeared at 1045 8C and 9 kbar, and indicated that F-rich phlogopite can be a stable phase in the
127
sapphirine þ quartz stability field in appropriate bulk compositions by the fractionation of available F into biotite during partial dehydration melting. This experimental result was supported by Motoyoshi & Hensen (2001), who analysed high fluorine contents (up to 8.2 wt% F) of phlogopites of pelitic granulite and quartz-free garnet-granulite from the Mt. Riiser-Larsen area and documented that very fluorine-rich phlogopite (6–8 wt% F) can be stable in UHT mineral assemblages with aluminous orthopyroxene, osumilite, pyrope-rich garnet, and sapphirine þ quartz. Similarly, Tsunogae et al. (2000) analysed high fluorine (up to 2.6 wt% F) and chlorine (up to 1.45 wt% Cl) contents of amphibole minerals in mafic granulites from Tonagh Island, and showed that amphibole minerals containing halogen elements could also be stable at UHT metamorphic temperatures. Subsequently, those workers considered that the H2O component in mica and amphibole minerals was replaced by a fluorine component at a temperature that is usually sufficient to melt mica and/or amphibole minerals in ordinary continental crust. These studies strongly demonstrate that such replacement may have played an important role in the very small degree of crustal melting during UHT metamorphism in the Napier Complex. For the P–T evolution of the Napier Complex, Motoyoshi & Hensen (1989) suggested a counterclockwise P– T path based on the detailed petrography of sapphirine þ quartz þ orthopyroxene symplectite, probably after cordierite, from the Mt. Riiser-Larsen area, whereas Osanai et al. (2001b) described sapphirine-consuming and osumilite-producing reactions from Bunt Island and suggested isothermal decompression in part of the clockwise P–T path. Tsunogae et al. (2001, 2002) also suggested that the CO2 isochores estimated by fluid inclusions in sapphirine and quartz from Tonagh Island intersect the counterclockwise P –T trajectory at around 6– 9 kbar at 1100 8C, whereas Tsunogae et al. (2003) described two types of CO2-rich fluid inclusions in garnet from Bunt Island (high-density and early entrapped inclusion, and low-density and late entrapped inclusion) and proposed a clockwise P–T evolution from 10 kbar at 1050 8C to ,7 kbar at ,950 8C. From other areas of the Napier Complex, both clockwise (Ellis 1987; Harley 1989) and counterclockwise (Sandiford & Powell 1986a, b; Hollis & Harley 2002) P–T –t paths have been also proposed, as summarized by Harley (2004). There is a wealth of textural descriptions of the post-peak stage as described above, but the prograde history of the Napier UHT metamorphism has been largely obliterated by the high-temperature conditions. This may partly result in the contrasting views as shown above. On the other hand, Hokada (1999) and Hokada et al. (2008) described two
128
Table 3. Age data estimated as metamorphic ages of the Napier Complex measured by the JARE geology group Sample locality
Method Sm –Nd bulk-rock
Tonagh Island
U –Pb (SHRIMP) mineral (margin of zircon)
Mt. Riiser-Larsen Mt. Riiser-Larsen
40
Mt. Riiser-Larsen
Ar – 39Ar bulk-rock U –Th–Pb (CHIME) mineral (zircon) U –Th–Pb (CHIME) mineral (monazite) U –Th–Pb (CHIME) mineral (zircon) U –Th–Pb (CHIME) mineral (monazite) U –Th–Pb (CHIME) mineral (zircon) U –Th–Pb (CHIME) mineral (monazite) Sm –Nd bulk-rock and mineral
Tonagh Island
U –Pb (SHRIMP) mineral (zircon) U –Pb (SHRIMP) mineral (monazite) U –Th–Pb (CHIME) mineral (zircon) Sm –Nd bulk-rock and mineral
Mt. Riiser-Larsen Christmas Point Tonagh Island Tonagh Island Mt. Riiser-Larsen
Sm –Nd bulk-rock and mineral U –Th–Pb (CHIME) mineral (xenotime) U –Pb (SIMS) mineral (margin of zircon) U –Pb (SHRIMP) mineral (margin of zircon) U –Pb (SHRIMP) mineral (zircon)
Mt. Riiser-Larsen
Age (Ga)
Reference
Felsic gneiss Mafic gneiss Felsic gneiss Felsic gneiss Felsic gneiss Felsic gneiss Felsic gneiss Felsic gneiss Felsic gneiss Felsite Felsic gneiss Felsic gneiss Mafic gneiss Garnet gneiss Felsic gneiss Felsic gneiss Felsic gneiss Alminous gneiss Felsic gneiss Garnet gneiss Felsic gneiss Pegmatite Felsic gneiss Felsic mylonite Felsic gneiss
2.56 2.60 2.66 2.55– 2.44 2.0 2.41 2.44 2.6– 2.3 2.6– 2.1 2.6– 2.4 2.5– 2.4 2.38 2.38, 2.30 2.36 2.83– 2.80, 2.64 – 2.44 2.46– 2.39 2.9– 2.8, 2.6– 2.3 1.87 1.90 1.56 2.20 2.17 2.47– 2.45 2.55– 2.47 2.52– 2.45
Tainosho et al. (1994, 1997) Shiraishi et al. (1997) Takigami et al. (1998) Asami et al. (1998) Hokada (1999)
Suzuki (2000)
Owada et al. (2001, 2003) Suzuki et al. (2001) Grew et al. (2002) Carson et al. (2002) Crowe et al. (2002) Hokada et al. (2003)
H. ISHIZUKA
Mt. Pardoe
Lithology
Mt. Riiser-Larsen Beaver Island Reference Peak Mt. Riiser-Larsen Mt. Riiser-Larsen
Bunt Island
U –Pb (SHRIMP) mineral (zircon) U –Pb (SHRIMP) mineral (monazite) U –Pb (conventional) mineral (zircon) U –Pb (conventional) mineral (rutile) Rb –Sr mineral (phlogopite) Sm –Nd bulk-rock and mineral Lu –Hf bulk-rock and mineral
Granulite Granulite Quartz gneiss Skarn Felsic gneiss Felsic gneiss Felsic gneiss Paragneiss Paragneiss Psammitic gneiss Mafic gneiss Granitic gneiss Granitic gneiss Granitic gneiss Felsic gneiss Granulite Granulite Granulite Layered gneiss
2.40 2.44 2.42 2.41 2.43 2.42 2.43 2.49– 2.48 2.47– 2.45 2.36 2.38 2.38 2.51– 2.47 2.47 2.44 1.5 1.85 1.85 2.40
Asami et al. (2002)
Hokada et al. (2004) Suzuki et al. (2006)
Miyamoto et al. (2006)
Choi et al. (2006)
JARE STUDIES IN THE NAPIER COMPLEX
Howard Hills
U –Th–Pb (CHIME) mineral (monazite) U –Th–Pb (CHIME) mineral (zircon) U –Th–Pb (CHIME) mineral (zircon) U –Th–Pb (CHIME) mineral (monazite) U –Th–Pb (CHIME) mineral (monazite) U –Th–Pb (CHIME) mineral (monazite) U –Th–Pb (CHIME) mineral (xenotime) U –Pb (SHRIMP) mineral (zircon) U –Th–Pb (CHIME) mineral (zircon) Sm –Nd bulk-rock and mineral
129
130
H. ISHIZUKA
types of reaction products after sapphirine þ quartz in the Mt. Riiser-Larsen area such as orthopyroxene þ sillimanite in the western area and orthopyroxene or cordierite in the eastern area of the shear zone (RLMSZ), and interpreted it as showing a slight difference in pressure conditions during the isobaric cooling between the western (higher-P) and eastern (lower-P) areas. This implies that the structural movement responsible for the formation of the shear zone could juxtapose rocks with different path of isobaric cooling; that is, rocks derived from different depths. Experimental studies have been applied to examine the P– T conditions of the Napier UHT metamorphism and related magmatism (Motoyoshi et al. 1993; Sakai & Kawasaki 1997; Kawasaki & Motoyoshi 2000, 2006, 2007; Hokada & Arima 2001; Sato & Kawasaki 2002; Sato et al. 2004, 2006). Motoyoshi et al. (1993) carried out a highpressure experiment in the K2O– MgO–Al2O3 – SiO2 (KMAS) system to determine the stability and phase relation of osumilite, and showed that (1) Mg-osumilite breaks down to enstatite, sillimanite, K-feldspar and quartz at pressures between 11 and 12 kbar over a temperature range of 950– 1100 8C with no appreciable dP/dT slope, and (2) the reactions osumilite ¼ cordierite þ enstatite þ K-feldspar þ quartz and osumilite ¼ sapphirine þ enstatite þ K-feldspar þ quartz limit the osumilite stability at lower and higher temperatures, respectively. Consequently, they suggested that in the Napier Complex an inferred ‘isograd’, marking the disappearance of osumilite, is the trace of a nearly isobaric surface at a palaeo-depth of c. 35 km, beyond which osumilite breaks down to orthopyroxene, sillimanite, K-feldspar and quartz. On the other hand, to study the origin of the (A-type?) granite that occurs as several small bodies or subconcordant veins in relatively lowtemperature regions in the Napier Complex (e.g. Napier Mountain and Tange Promontory; Sheraton et al. 1987), Hokada & Arima (2001) carried out a melting experiment on a mixture of feldspar (antiperthite) þ quartz þ orthopyroxene separated from the felsic gneiss of the Mt. Riiser-Larsen area at 1000– 1150 8C and 10 kbar under dry conditions, and showed that a granitic glass (,10 vol.%) with a chemical composition of A-type granite was detected along boundaries between lamella-free plagioclase and quartz grains in the run at 1150 8C. Therefore, they proposed that some genetic links might exist between UHT metamorphism and A-type granite magmatism in the Napier Complex. The outcome of JARE geological investigations also includes determinations of metamorphic ages of the Napier Complex by means of isotope dating methods for Sm –Nd bulk-rock and minerals,
Lu –Hf bulk-rock and minerals, 40Ar – 39Ar bulkrock, U –Pb SHRIMP and SIMS mineral, and chemical dating by CHIME (Table 3), showing that the data are scattered from 2.9 to 1.6 Ga. Of these, the U– Pb SHRIMP and SIMS mineral data include the ages 2.66 and 2.55– 2.44 Ga (Shiraishi et al. 1997), 2.83–2.80, 2.64–2.44 and 2.46 – 2.39 Ga (Suzuki 2000), 2.47 –2.45 Ga (Carson et al. 2002), 2.55–2.47 Ga (Crowe et al. 2002), 2.52 –2.45 Ga (Hokada et al. 2003), 2.49–2.48 Ga (Hokada et al. 2004) and 2.51 –2.47 and 2.47 Ga (Suzuki et al. 2006). From these data, there are two interpretations on the timing of the UHT Napier metamorphism: c. 2.55 Ga (Crowe et al. 2002) and c. 2.51 –2.45 Ga (Carson et al. 2002; Hokada et al. 2004; Suzuki et al. 2006). Asami et al. (2002) reported CHIME ages of 2.40 – 2.44 Ga and interpreted these ages to be the peak event of UHT metamorphism. In addition to the JARE studies, U –Pb SHIRIMP and SIMS zircon data also suggest two further ages: c. 2.59 – 2.55 Ga (Harley et al. 2001; Kelly & Harley 2005) and c. 2.50 –2.45 Ga (e.g. Grew 1998). Harley & Black (1997) proposed c. 2.84 Ga for the peak age of the Napier UHT metamorphism, but this age was later considered by Kelly & Harley (2005) to relate to a low-P/high-T type of metamorphism and not to the c. 8–11 kbar UHT metamorphic event in the Napier Complex. Younger ages in Table 3 were interpreted as reset ages caused by local thermal effects such as intrusion or pegmatite activities, or by local structural effects such as the development of shear zones.
Discussion As summarized above, the geological studies of JARE have provided additional observations and data, and some new aspects and interpretations for the geology, protolith and metamorphism of the Napier Complex. However, there are still several topics of considerable debate, such as the role and age of TTG protolith, and the origin and timing of UHT metamorphism, which are briefly considered below.
Role and age of TTG protolith The TTG protolith as reported from the Mt. RiiserLarsen area (Suzuki et al. 1999) was also described from other areas of the Napier Complex (Sheraton & Black 1983; Sheraton et al. 1987). The origin of the modern TTG suite has been explained as the product of magmatic processes, namely melting of basaltic source rocks (low-Mg amphibolites) principally occurring in a subduction zone (e.g. Foley et al. 2002). A recent study of seismic
JARE STUDIES IN THE NAPIER COMPLEX
experiments demonstrated that TTG varieties (tonalitic rocks), derived from the anatexis of the differentiated basaltic lower crust (Tatsumi 2000), exist in the middle to lower crust of the Mariana intra-oceanic island arc (Takahashi et al. 2007). This is supported by the occurrence of obducted tonalitic rocks at the northern tip (the Tanzawa Mountain) of the Izu arc where the arc collides with the Japan arc (Kawate & Arima 1998). These facts are indicative of the presence of a ‘modern intra-oceanic island arc-like’ tectonic setting that may have played an important role in the formation of Napier TTG protoliths. However, the Archaean TTG suite is slightly different from the modern variety in its geochemistry (Martin 1994; Luais & Hawkesworth 1994), such as high chondrite-normalized (La/Yb)N ratio and low YbN content, and depleted heavy REE (HREE) in the Archaean TTG suite, which has also been described in the Napier TTG protolith (Sheraton et al. 1987; Suzuki et al. 1999). These geochemical features of the Archaean TTG suite require a source material with residual amphibole and/or garnet, because of the strong partitioning of HREE into these minerals (e.g. Martin 1994). This difference in geochemistry between the Archaean and modern TTG suites is considered to result from the difference of physical conditions for the production of TTG magma, such as a difference in the geothermal gradient at a subduction zone from the high T/P gradient during Archaean time to the low T/P variety at present; that is, at similar depths the high T/P geothermal gradient favours the stability of garnet and amphibole (e.g. Condie 1997). The assumption that the Archaean subducted plate was relatively young (,30 Ma) and warm is a major basis for this model. There is, however, no field evidence for this model, and it is still debated. Therefore, further study on the Napier TTG protolith, especially study of its field occurrence and field relations with surrounding rocks, may provide constraints for understanding the birth of Archaean continental crust. In this respect, the regional field relations need to be investigated in more detail in the Napier Complex. The age of protoliths is another subject requiring further research. As described above, there are at least four clusters of protolith ages (c. 3.8, 3.3, 3.0 and 2.6 Ga) reported from the Napier Complex, all of which were U –Pb SHRIMP or SIMS zircon ages from the TTG or granitic protolith (now represented by felsic gneisses or orthogneisses). These ages possibly indicate the magmatic episodes by which the protoliths were generated. However, they do not always indicate the formation ages of juvenile (mantle-derived) crust; that is, there is a possibility that the protoliths were derived from preexisting materials such as basalts or sediments by
131
remelting processes. Distinction between juvenile crust and remelting crust is essential to evaluate the rate of crustal growth during Earth’s history (Condie 1997), and it can be examined by means other than the U –Pb isotope method alone. Recent progress in U –Pb zircon isotope geochemistry combined with oxygen and Lu –Hf isotope systematics analysed using laser-ablation multicollector inductively coupled plasma-mass spectrometry (MC-ICP-MS) shows that (1) the model age of Lu –Hf isotopes is useful to examine the formation age of juvenile crust, and (2) the oxygen isotope can distinguish mantle-derived magma (low d18O, e.g. ,6.5‰) from remelting magma (high d18O, e.g. .6.5‰), because higher d18O values reflect a component of 18O-enriched, supracrustal material in the magma from which the zircon precipitated, which could include recycled sedimentary rocks or hydrothermally altered oceanic crust (Cavosie et al. 2005; Hawkesworth & Kemp 2006; Kemp et al. 2006). It is, therefore, important that the protolith ages of the Napier Complex are re-examined by using in situ analysis of U –Pb, oxygen, and Lu –Hf zircon isotopes, which will provide important information that will help in modelling the origin and evolution of the Napier continental crust.
Origin and timing of UHT metamorphism The origin of UHT metamorphism, such as heat source and related tectonics, has a key role in understanding the process that stabilized the continental crust. In particular, the fact that the Napier Complex showed little sign of melting at temperature up to 1130 8C provides a constraint on the thermal structure of the Archaean continental crust and related behaviour of volatile components, because the temperature at similar depths of the modern lower continental crust is assumed to be 800–900 8C. On the basis of the thermal as well as pressure gradations, and the P –T evolution, several models have been proposed for the origin of the Napier UHT metamorphism, such as an asthenosphere heat advection model related to continental collision and thickening (Ellis 1987; Harley 1989, 2004; Arnold et al. 2001) or a magma intrusion model (Sandiford & Powell 1986b; Motoyoshi & Hensen 1989; Hensen & Motoyoshi 1992) or a combination model (Suzuki et al. 2006). There is, however, little field evidence supporting these speculative models, and, as described above the P –T evolution is still a matter of debate. In this respect, very recently, Soyama & Ishizuka (2007a, b) described a UHT granulite sample from the Mt. Riiser-Larsen area, which is composed of osumilite, sapphirine, garnet, orthopyroxene (Al2O3 9.4– 14.9 wt%), sillimanite, K-feldspar, biotite, spinel, and
132
H. ISHIZUKA
quartz. They examined the microstructures using a combination of confocal micro-Laser Raman Spectroscopy (LRS) and electron microprobe analysis (EMPA) to find the presence of kyanite þ orthopyroxene (Al2O3 5.0–7.4 wt%) association occurring as an inclusion in osumilite. This finding is critical as the P– T conditions change from the kyanite þ low-Al orthopyroxene stability field to the osumilite þ sillimanite þ high-Al orthopyroxene stability field (lower-P and higher-T field) during the prograde clockwise P –T path in the Riiser-Larsen area. Furthermore, the fact that this sample was collected from the eastern region of the RLMSZ is consistent with the regional subdivision of the Napier Complex into a clockwise and counterclockwise P–T –t path metamorphic unit as shown by Toyoshima et al. (2008, Fig. 5). It is, therefore, suggested that detailed analyses of P –T conditions and evolution of the UHT metamorphism in the Napier Complex require further detailed microstructural studies using new techniques such as LRS. The precise timing of the UHT Napier metamorphism is also controversial at present, and on the basis of the U – Pb SHRIMP and SIMS zircon ages, there are at least two contrasting ages as described above; c. 2.59 –2.55 Ga and c. 2.50– 2.45 Ga. The use of cathodoluminescence (CL) and backscattered electron (BSE) image analyses, scanning electron microscope (SEM) imaging, and analyses of trace elements such as Th and U have led to attempts that link zoning patterns, internal textures and compositional features of zircon grains in distinguishing igneous or metamorphic origins (Carson et al. 2002; Hokada et al. 2004; Kelly & Harley 2005). This technique has been extensively applied for the dating of protoliths of the Napier Complex, as shown in Table 2. It is, however, not always entirely clear what kinds of zircon textures have been formed during metamorphism; for instance, by growth of new zircon or recrystallization of pre-existing zircon (e.g. Schaltegger et al. 1999; Kelly et al. 2002). Consequently, it is unclear which minerals were in equilibrium with the zircon, and what kinds of metamorphic reactions result in the production of zircon. These problems seem to have largely contributed to the contrasting views on the timing of the UHT metamorphism in the Napier Complex. In this regard, the analyses of zirconium for constituent minerals such as garnet, osumilite, cordierite and sillimanite carried out by Fraser et al. (1997), Pan (1997) and Degeling et al. (2001) are important to obtain a constraint on zircon formation. For example, Degeling et al. (2001) proposed that (1) in the decompression reaction garnet (Zr 24.75 ppm) þ sillimanite (0.08 ppm) þ quartz ¼ cordierite (1.57 ppm), Zr released during
garnet breakdown cannot be incorporated in the cordierite structure, resulting in zircon nucleation and growth, and (2) for the reaction garnet (Zr 62.10 ppm) þ biotite þ sillimanite þ quartz ¼ osumilite (38.71 ppm) þ orthopyroxene (31.80 ppm) þ spinel (0.21 ppm) þ magnetite (0.09 ppm), no new zircon growth takes place. On the other hand, recent studies demonstrate that an analysis of zircon for trace or rare earth elements combined with CL and SEM images, and with U–Pb SHRIMP or SIMS dating is a potentially powerful chemical approach to link zircon growth and recrystallization to metamorphic events (Harley et al. 2001; Hokada & Harley 2004; Kelly & Harley 2005). Hokada & Harley (2004) analysed REE in zircon and coexisting garnet from the UHT gneiss of the Mt. Riiser-Larsen area, and, based on comparisons with recent estimates of equilibrium zircon–garnet REE distribution coefficients (Harley et al. 2001; Rubatto 2002; Whitehouse & Platt 2003), they inferred that the inner core of zircon did not grow with the garnet but grew within a garnet-absent melt that was then injected into the gneiss during the time interval 2.50–2.47 Ga in the waning stages of the Napier UHT metamorphism. Also, Kelly & Harley (2005) used a similar approach for samples from the Gage Ridge, Proclamation Island, Dallwitz Nunatak and Zircon Point, and proposed c. 2.55 Ga as the timing of the Napier UHT metamorphism. Consequently, they suggested that the high proportion of published zircon U–Pb data with ages between c. 2.49 and 2.45 Ga reflects late, post-peak zircon growth and does not date the timing of the peak Napier UHT metamorphism. These new approaches can, therefore, provide important clues as to when and how the zircon has been crystallized, which promises new vision for future study of the Napier Complex.
Conclusions The geological studies of JARE in the Napier Complex can be summarized as follows. 1.
2.
Field studies revealed that the Napier Complex comprises layered and massive gneiss units with nine stages of deformation history. The boundary between the units is transitional in the Mt. Riiser-Larsen area, but closely associated with the shear zone on Tonagh Island. Geochemical studies showed that (a) garnet– sillimanite gneisses and garnet-rich felsic gneisses were derived from mudstone and sandstone, respectively, both enriched in MgO, Cr and Ni, (b) orthopyroxene felsic gneisses have a close REE affinity with Archaean TTG, (c) basic gneisses were derived from LREE-enriched or -depleted basalts, and
JARE STUDIES IN THE NAPIER COMPLEX
3.
4.
(d) meta-ultramafic rocks are comparable with komatiite and related depleted mantle peridotite. This suite of protoliths is reminiscent of Archaean greenstone–granite belts. Precise analyses of physical conditions of the Napier UHT metamorphism by using new approaches indicated that the maximum metamorphic temperature reached 1130 8C. P–T evolution was examined by the analyses of reaction textures combined with fluid inclusion studies, and suggested both clockwise (Bunt Island) and counterclockwise (Mt. RiiserLarsen and Tonagh Island) P– T–t paths. U –Pb SHRIMP and SIMS zircon ages revealed three stages of protolith formation at c. 3.3, 3.0 and 2.6 Ga. The timing of peak UHT metamorphism is still unresolved between c. 2.55 and c. 2.51 –2.45 Ga.
On the basis of these results, further studies in the Napier Complex, such as regional structural and tectonic characterization based on geological field studies, microstructural studies that focus on the pre-UHT metamorphic imprints using advanced techniques such as LRS, and radiogenic and stable isotopes coupled with trace element imaging, should be carried out. The results of these studies will advance our understanding of the role of TTG protolith, causes and consequences of UHT crustal metamorphism and will help in modelling the Archaean continental evolution more accurately. I gratefully acknowledge all members of the JARE geology group, especially the Nishi–Higashi group, with whom I enjoyed frank discussions during the course of the study. I thank C. J. Carson, T. Hokada, M. Satish-Kumar and Y. Zhao for careful and constructive reviews of the manuscript.
References A RNOLD , J., J ACOBY , W. R., S CHMELING , H. & S CHOTT , B. 2001. Continental collision and the dynamic and thermal evolution of the Variscan orogenic crustal root—numerical models. Journal of Geodynamics, 31, 273–291. A SAMI , M., S UZUKI , K., G REW , E. S. & A DACHI , M. 1998. CHIME ages for granulite from the Napier Complex, East Antarctica. Polar Geoscience, 11, 172–199. A SAMI , M., S UZUKI , K. & G REW , E. S. 2002. Chemical Th– U– total Pb dating by electron microprobe analysis of monazite, xenotime and zircon from the Archean Napier Complex, East Antarctica: evidence for ultra-high-temperature metamorphism at 2400 Ma. Precambrian Research, 114, 249– 275. B LACK , L. P., J AMES , P. R. & H ARLEY , S. L. 1983. The geochronology, structure and metamorphism of early Archaean rocks at Fyfe Hills, Enderby Land, Antarctica. Precambrian Research, 21, 197–222.
133
B LACK , L. P., F ITZGERALD , J. D. & H ARLEY , S. L. 1984. Pb isotopic composition, color, and microstructure of monazite from a polymetamorphic rock in Antarctica. Contributions to Mineralogy and Petrology, 85, 141– 148. B LACK , L. P., W ILLIAMS , I. S. & C OMPSTON , W. 1986a. Four zircon age from one rock: the history of a 3930-Ma-old granulite from Mount Sones, Enderby Land, Antarctica. Contributions to Mineralogy and Petrology, 94, 427–437. B LACK , L. P., S HERATON , J. W. & J AMES , P. R. 1986b. Late Archaean granite of the Napier Complex, Enderby Land, Antarctica: a comparison of Rb–Sr, Sm–Nd and U– Pb isotopic systematics in a complex terrain. Precambrian Research, 32, 343– 368. C ARSON , C. J., A GUE , J. J. & C OATH , C. D. 2002. U–Pb geochronology from Tonagh Island, East Antarctica: implications for the timing of ultra-high temperature metamorphism of the Napier Complex. Precambrian Research, 116, 237–263. C AVOSIE , A. J., V ALLEY , J. W., W ILDE , S. A. & EDINBURGH ION MICROPROBE FACILITY (EIMF). 2005. Magmatic d18O in 4400–3900 Ma detrital zircons: a record of the alteration and recycling of crust in the Early Archean. Earth and Planetary Science Letters, 235, 663–684. C HOI , S. H., M UKASA , S. B., A NDRONIKOV , A. V., O SANAI , Y., H ARLEY , S. L. & K ELLY , N. M. 2006. Lu–Hf systematics of the ultra-high temperature Napier Metamorphic Complex in Antarctica: evidence for the early Archean differentiation of Earth’s mantle. Earth and Planetary Science Letters, 246, 305–316. C ONDIE , K. C. 1994. Archean Crustal Evolution. Elsevier, Amsterdam. C ONDIE , K. C. 1997. Plate Tectonics and Crustal Evolution. Butterworth– Heinemann, Oxford. C ONDIE , K. C. & W RONKIEWICZ , D. J. 1990. The Cr/Th ratio in Precambrian pelites from the Kaapvaal craton as an index of craton evolution. Earth and Planetary Science Letters, 97, 256–267. C ROWE , W. A., O SANAI , Y., T OYOSHIMA , T., O WADA , M., T SUNOGAE , T. & H OKADA , T. 2002. SHRIMP geochronology of a mylonite zone on Tonagh Island: characterization of the last high-grade tectonothermal event in the Napier Complex, East Antarctica. Polar Geoscience, 15, 17– 36. D ALLWITZ , W. B. 1968. Coexisting sapphirine and quartz in granulite from Enderby Land, Antarctica. Nature, 219, 476–477. D EGELING , H., E GGINS , S. & E LLIS , D. J. 2001. Zr budgets for metamorphic reactions, and the formation of zircon from garnet breakdown. Mineralogical Magazine, 65, 749– 758. E LLIS , D. J. 1980. Osumilite– sapphirine– quartz granulite from Enderby Land, Antarctica: P– T conditions of metamorphism, implications for garnet– cordierite equilibria and the evolution of the deep crust. Contributions to Mineralogy and Petrology, 74, 201– 210. E LLIS , D. J. 1987. Origin and evolution of granulites in normal and thickened crusts. Geology, 15, 167– 170. F OLEY , S., T IEPOLO , M. & V ANNUCCI , R. 2002. Growth of early continental crust controlled by melting of amphibolite in subduction zones. Nature, 417, 837– 840.
134
H. ISHIZUKA
F RASER , G., E LLIS , D. J. & E GGINS , S. 1997. Zirconium abundance in granulite-facies minerals, with implications for zircon geochronology in high-grade rocks. Geology, 25, 607–610. G REW , E. S. 1980. Sapphirine þ quartz association from Archaean rocks in Enderby Land, Antarctica. American Mineralogist, 65, 821–836. G REW , E. S. 1982a. The Antarctic Margin. In: N AIRN , A. E. M. & S TEHLI , F. G. (eds) The Ocean Basins and Margins, Vol. 6. Plenum, New York, 679– 755. G REW , E. S. 1982b. Osumilite in the sapphirine–quartz terrane of Enderby Land, Antarctica: implications for osumilite petrogenesis in the granulite facies. American Mineralogist, 67, 762–787. G REW , E. S. 1998. Boron and beryllium mineral in granulite-facies pegmatites and implications of beryllium pegmatites for the origin and evolution of the Archaean Napier Complex of East Antarctica. In: M OTOYOSHI , Y. & S HIRAISHI , K. (eds) Origin and Evolution of Continents. Memoirs of the National Institute of Polar Research, Special Issue, 53, 74– 92. G REW , E. S. & M ANTON , W. I. 1979. Archean rocks in Antarctica: 2.5 billion-year uranium–lead ages of pegmatites in Enderby Land. Science, 206, 443– 445. G REW , E. S., S UZUKI , K. & A SAMI , M. 2002. CHIME ages of xenotime, monazite and zircon from beryllium pegmatites in the Napier Complex, Khmara Bay, Enderby Land, East Antarctica. Polar Geoscience, 14, 99– 118. H ARLEY , S. L. 1985. Garnet– orthopyroxene bearing granulites from Enderby Land, Antarctica: metamorphic pressure– temperature– time evolution of the Archaean Napier Complex. Journal of Petrology, 26, 819– 856. H ARLEY , S. L. 1987. A pyroxene-bearing meta-ironstone and other pyroxene granulite from Tonagh Island, Enderby Land, Antarctica: further evidence for very high temperature (.980 8C) Archaean regional metamorphism in the Napier Complex. Journal of Metamorphic Geology, 5, 341–356. H ARLEY , S. L. 1989. The origin of granulites: a metamorphic perspective. Geological Magazine, 126, 215– 247. H ARLEY , S. L. 1998. The occurrence and characterisation of ultrahigh-temperature crustal metamorphism. In: T RELOAR , P. J. & O’B RIEN , P. J. (eds) What Drives Metamorphism and Metamorphic Reactions? Geological Society, London, Special Publications, 138, 81–107. H ARLEY , S. L. 2004. Extending our understanding of ultrahigh temperature crustal metamorphism. Journal of Mineralogical and Petrological Sciences, 99, 140– 158. H ARLEY , S. L. & B LACK , L. P. 1997. A revised Archaean chronology for the Napier Complex, Enderby Land, from SHRIMP ion-microprobe studies. Antarctic Science, 9, 74–91. H ARLEY , S. L. & H ENSEN , B. J. 1990. Archaean and Proterozoic high-grade terranes of East Antarctica (40–80 8E): a case study of diversity in granulite facies metamorphism. In: A SHWORTH , J. R. & B ROWN , M. (eds) High-temperature Metamorphism and Crustal Anatexis. Unwin Hyman, London, 320– 370.
H ARLEY , S. L. & M OTOYOSHI , Y. 2000. Al zoning in orthopyroxene in a sapphirine quartzite: evidence for . 1120 8C UHT metamorphism in the Napier Complex, Antarctica, and implications for the entropy of sapphirine. Contributions to Mineralogy and Petrology, 138, 293–307. H ARLEY , S. L., K INNY , O. D., S NAPE , I. & B LACK , L. P. 2001. Zircon chemistry and the definition of events in Archaean granulite terrains. In: C ASSIDY , K. F., D UNPHY , J. M. & V AN K RANENDONK , M. J. (eds) Extended Abstracts of the 4th International Archaean Symposium. AGSO-Geoscience Australia Record, Canberra, A.C.T., 511–513. H AWKESWORTH , C. J. & K EMP , A. I. S. 2006. Using hafnium and oxygen isotopes in zircons to unravel the record of crustal evolution. Chemical Geology, 226, 144– 162. H ENSEN , B. J. & M OTOYOSHI , Y. 1992. Osumiliteproducing reactions in high-temperature granulite from the Napier Complex, East Antarctica: tectonic implications. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Science. Terra, Tokyo, 87–92. H ENSEN , B. J. & O SANAI , Y. 1994. Experimental study of dehydration melting of F-bearing biotite in model pelitic compositions. Mineralogical Magazine, 58A, 410–411. H OKADA , T. 1999. Thermal evolution of the ultrahightemperature metamorphic rocks in the Archaean Napier Complex, East Antarctica. PhD thesis, Graduate University for Advanced Studies, Tokyo. H OKADA , T. 2001. Feldspar thermometry in ultrahightemperature metamorphic rocks: evidence of crustal metamorphism attaining 1100 8C in the Archaean Napier Complex, East Antarctica. American Mineralogist, 86, 932 –938. H OKADA , T. & A RIMA , M. 2001. Experimental constraints on partial melting under UHT and dry conditions of quartzo-feldspathic rock in the Napier Complex, East Antarctica. Polar Geoscience, 14, 39–52. H OKADA , T. & H ARLEY , S. L. 2004. Zircon growth on UHT leucosome: constraints from zircon–garnet rare earth elements (REE) relations in Napier Complex, East Antarctica. Journal of Mineralogical and Petrological Sciences, 99, 180–190. H OKADA , T. & S UZUKI , S. 2006. Feldspar in felsic orthogneiss as indicator for UHT crustal processes. Journal of Mineralogical and Petrological Sciences, 101, 260– 264. H OKADA , T., O SANAI , Y., T OYOSHIMA , T., O WADA , M., T SUNOGAE , T. & C ROWE , A. W. 1999. Geology and metamorphism of sapphirine-bearing aluminous gneisses from Tonagh Island in the Napier Complex, East Antarctica. Polar Geoscience, 12, 49– 70. H OKADA , T., M ISAWA , K., S HIRAISHI , K. & S UZUKI , S. 2003. Mid to late Archaean (3.3–2.5 Ga) tonalitic crustal formation and high-grade metamorphism at Mt. Riiser-Larsen, Napier Complex, East Antarctica. Precambrian Research, 127, 215–228. H OKADA , T., M ISAWA , K., Y OKOYAMA , K., S HIRAISHI , K. & Y AMAGUCHI , A. 2004. SHRIMP and electron microprobe chronology of UHT metamorphism in the Napier Complex, East Antarctica: implication for
JARE STUDIES IN THE NAPIER COMPLEX zircon growth at .1000 8C. Contributions to Mineralogy and Petrology, 147, 1– 20. H OKADA , T., M OTOYOSHI , Y., S UZUKI , S., I SHIKAWA , M. & I SHIZUKA , H. 2008. Geodynamic evolution of Mt. Riiser-Larsen, Napier Complex, East Antarctica, with reference to the UHT mineral associations and their reaction relations. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East–West Gondwana Connection. Geological Society, London, Special Publications, 308, 255– 284. H OLLIS , J. & H ARLEY , S. L. 2002. New evidence for the peak temperature and the near-peak pressure– temperature evolution of the Napier Complex. In: G AMBLE , J. A., S KINNER , D. N. B., H ENRYS , S. & L YNCH , R. (eds) Antarctica at the Close of a Millennium. Proceedings Volume, 8th International Symposium on Antarctic Earth Sciences. Royal Society of New Zealand Bulletin, 35, 19–29. I SHIKAWA , M., H OKADA , T., I SHIZUKA , H., M IURA , H., S UZUKI , S., T AKADA , M. & Z WARTZ , D. P. 2000. Antarctic Geological Map Series, Sheet 37, Mount Riiser-Larsen, with explanatory text. National Institute of Polar Research, Tokyo. I SHIZUKA , H., M IURA , H., T AKADA , M., I SHIKAWA , M., Z WARTZ , D. P., S UZUKI , S. & H OKADA , T. 1997a. Report on the geological and geomorphological field party in the Mt. Riiser-Larsen area, Enderby Land, 1996–97 (JARE-38). Antarctic Record, 41, 743– 777. I SHIZUKA , H., I SHIKAWA , M., S UZUKI , S. & H OKADA , T. 1997b. The Napier Complex in East Antarctica—The oldest craton and the ultra-high temperature metamorphism. Journal of Geological Society of Japan, 103, XXIX– XXX [in Japanese]. I SHIZUKA , H., I SHIKAWA , M., H OKADA , T. & S UZUKI , S. 1998. Geology of the Mt. Riiser-Larsen area of the Napier Complex, Enderby Land, East Antarctica. Polar Geoscience, 11, 154– 171. I SHIZUKA , H., Y OSHIDA , T. & S UZUKI , S. 1999. Peak metamorphic temperature of the Napier Complex deduced from bulk rock and pyroxene compositions of quartz– magnetite rocks. In: Abstracts of the 19th Symposium on Antarctic Geosciences, Tokyo, 41. I SHIZUKA , H., S UZUKI , S. & N AKAMURA , A. 2002. Peak temperatures of ultra-high temperature metamorphism of the Napier Complex, Enderby Land, East Antarctica, as deduced from porphyroclastic pyroxenes of meta-ultramafic rocks. Polar Geoscience, 15, 1– 16. K AMENEV , E. N. 1975. The geology of Enderby Land. Academy of Science of the USSR Committee on Antarctic Research Report, 14. K AWASAKI , T. & M OTOYOSHI , Y. 2000. High-pressure and high-temperature phase relations of an orthopyroxene granulite from McIntyre Island, Enderby Land, East Antarctica. Polar Geoscience, 13, 114–134. K AWASAKI , T. & M OTOYOSHI , Y. 2006. Experimental constraints on the decompressional P –T paths of Rundva˚gshetta granulite, Lu¨tzow-Holm Complex, East Antarctica. In: F U¨ TTERER , D. K., D AMASKE , D., K LEINSCHMIDT , G., M ILLER , H. & T ESSENSOHN , F. Antarctica: Contributions to Global Earth Science. Springer, Berlin, 23– 36.
135
K AWASAKI , T. & M OTOYOSHI , Y. 2007. Solubility of TiO2 in garnet and orthopyroxene: Ti thermometer for ultrahigh-temperature granulite. In: C OOPER , A. K. & R AYMOND , C. ISAES Editorial Team (eds) Antarctica: A Keystone in a Changing World. US Geological Survey Open-File Report, 2007-1047; Short Research Paper, 038, doi: 10.3133/of20071047.srp038. K AWATE , S. & A RIMA , M. 1998. Petrogenesis of the Tanzawa plutonic complex, central Japan: exposed felsic middle crust of the Izu –Bonin–Mariana arc. Island Arc, 7, 342– 358. K ELLY , N. M. & H ARLEY , S. L. 2005. An integrated microtextural and chemical approach to zircon geochronology: refining the Archaean history of the Napier Complex, east Antarctica. Contributions to Mineralogy and Petrology, 149, 57– 84. K ELLY , N. M., C LARK , G. L. & F ANNING , C. M. 2002. A two-stage evolution of the Neoproterozoic Rayner Structural Episode: new U– Pb SHRIMP constraints from the Oygarden Group, Kemp Land, East Antarctica. Precambrian Research, 116, 301– 330. K EMP , A. I. S., H AWKESWORTH , C. J., P ATERSON , B. A. & K INNY , P. D. 2006. Episodic growth of the Gondwana supercontinent from hafnium and oxygen isotopes in zircon. Nature, 429, 580– 583. L UAIS , B. & H AWKESWORTH , C. 1994. The generation of continental crust: An integrated study of crust forming processes in the Archaean of Zimbabwe. Journal of Petrology, 35, 43– 93. M AKIMOTO , H., A SAMI , M. & G REW , E. S. 1989. Some geological observations on the Archean Napier Complex at Mt. Riiser-Larsen, Amundsen Bay, Enderby Land. Proceedings of NIPR Symposium on Antarctic Geosciences, 3, 128–141. M ARTIN , H. 1994. The Archean gray gneisses and the genesis of continental crust. In: C ONDIE , K. C. (ed.) Archean Crustal Evolution, Elsevier, New York, 205– 259. M IYAMOTO , T., Y OSHIMURA , Y., S ATO , K., M OTOYOSHI , Y., D UNKLEY , D. J. & C ARSON , C. J. 2004. Occurrence of metamorphosed ultramafic rock and associated rocks in Howard Hills, Enderby Land, East Antarctica: Evidence of partial melting from geochemical and isotopic characteristics. Polar Geoscience, 17, 88– 111. M IYAMOTO , T., Y OSHIMURA , Y., D UNKLEY , D. J., M OTOYOSHI , Y. & C ARSON , C. J. 2006. Mineral ages for multi isotope system in phlogopite-bearing pyroxene granulite and felsic gneiss, the Howard Hills, Enderby Land, East Antarctica: Possible Proterozoic tectonothermal events in the Napier Complex. Polar Geoscience, 19, 152–172. M ORIWAKI , K. 1998. Activities of the Summer Party of the 39th Japanese Antarctic Research Expedition in 1997-1998. Antarctic Record, 42, 300–320 [in Japanese]. M OTOYOSHI , Y. 1998. Ultra-high temperature metamorphism of the Napier Complex, East Antarctica: a metamorphic study. Journal of Geological Society of Japan, 104, 794– 807. M OTOYOSHI , Y. & H ENSEN , B. J. 1989. Sapphirine – quartz– orthopyroxene symplectites after cordierite in the Archaean Napier Complex, Antarctica: evidence
136
H. ISHIZUKA
for a counterclockwise P– T path? European Journal of Mineralogy, 1, 467–471. M OTOYOSHI , Y. & H ENSEN , B. J. 2001. F-rich phlogopite stability in ultra-high-temperature metapelites from the Napier complex, East Antarctica. American Mineralogist, 86, 1404–1413. M OTOYOSHI , Y. & M ATSUEDA , H. 1984. Archean granulites from Mt. Riiser-Larsen in Enderby Land, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 33, 103– 125. M OTOYOSHI , Y. & M ATSUEDA , H. 1987. Corundum þ quartz association in Archean granulite-facies rock from Enderby Land, East Antarctica: preliminary interpretation. Proceedings of NIPR Symposium on Antarctic Geosciences, 1, 107–112. M OTOYOSHI , Y., H ENSEN , B. J. & M ATSUEDA , H. 1990. Metastable growth of corundum adjacent to quartz in a spinel-bearing quartzite from the Archaean Napier Complex, Antarctica. Journal of Metamorphic Geology, 8, 125– 130. M OTOYOSHI , Y., H ENSEN , B. J. & A RIMA , M. 1993. Experimental study of the high-pressure stability limit of osumilite in the system K2O– MgO–Al2O3 – SiO2: implications for high-temperature granulites. European Journal of Mineralogy, 5, 439–445. M OTOYOSHI , Y., M IURA , H., Y AMAUCHI , H. ET AL . 1999. Report on the geological and geomorphological field operation in the Amundsen Bay region, western Enderby Land, 1998-99 (JARE-40). Antarctic Record, 43, 534– 570. O SANAI , Y., T OYOSHIMA , T., O WADA , M., T SUNOGAE , T., H OKADA , T. & C ROWE , A. W. 1999. Geology of ultrahigh-temperature metamorphic rocks from Tonagh Island in the Napier Complex, East Antarctica. Polar Geoscience, 12, 1– 28. O SANAI , Y., T OYOSHIMA , T., O WADA , M., ET AL . 2001a. Antarctic Geological Map Series, Sheet 38, Tonagh Islands, with explanatory text. National Institute of Polar Research, Tokyo. O SANAI , Y., T OYOSHIMA , T., O WADA , M., T SUNOGAE , T., H OKADA , T., C ROWE , A. W. & K USACHI , I. 2001b. Ultrahigh temperature sapphirine–osumilite and sapphirine–quartz granulites from Bunt Island in the Napier complex, East Antarctica—Reconnaissance estimation of P–T evolution. Polar Geoscience, 14, 1–24. O WADA , M., O SANAI , Y. & K AGAMI , H. 1994. Isotopic equilibration age of Sm– Nd whole-rock system in the Napier Complex (Tonagh Island), East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 7, 122 –132. O WADA , M., O SANAI , Y., T OYOSHIMA , T., T SUNOGAE , T., H OKADA , T. & C ROWE , A. W. 1999. Petrography and geochemistry of mafic and ultramafic rocks from Tonagh Island in the Napier Complex, East Antarctica: a preliminary report. Polar Geoscience, 12, 87– 100. O WADA , M., O SANAI , Y., T SUNOGAE , T., T OYOSHIMA , T., H OKADA , T. & C ROWE , A. W. 2000. LREEenriched mafic gneiss and meta-ultramafic rock from Tonagh Island in the Napier Complex, East Antarctica. Polar Geoscience, 13, 86–102. O WADA , M., O SANAI , Y., T SUNOGAE , T., H AMAMOTO , T., K AGAMI , H., T OYOSHIMA , T. & H OKADA , T. 2001. Sm– Nd garnet ages of retrograde garnet bearing granulites from Tonagh Island in the Napier
Complex, East Antarctica: A preliminary study. Polar Geoscience, 14, 75–87. O WADA , M., O SANAI , Y., T OYOSHIMA , T., T SUNOGAE , T., H OKADA , T., C ROWE , A. W. & K AGAMI , H. 2003. Early Proterozoic tectonothermal events in the Napier Complex, East Antarctica: Implications for the formation of East Gondwana. Gondwana Research, 6, 231–240. P AN , Y. 1997. Zircon- and monazite-forming metamorphic reactions at Manitouwadge, Ontario. Canadian Mineralogist, 35, 105– 118. R AVICH , M. G. & K AMENEV , E. N. 1975. Crystalline Basement of the Antarctic Platform. Wiley, New York. R UBATTO , D. 2002. Zircon trace element geochemistry: partitioning with garnet and the link between U– Pb ages and metamorphism. Chemical Geology, 184, 123–138. S AKAI , S. & K AWASAKI , T. 1997. An experimental study of Fe– Mg partitionings between orthopyroxene and cordierite in the Mg-rich portion of the Mg3Al2Si3O12 – Fe3Al2Si3O12 system at atmospheric pressure: Calibration of its geothermometer for hightemperature granulite and igneous rocks. Proceedings of NIPR Symposium on Antarctic Geosciences, 10, 165–177. S ANDIFORD , M. & P OWELL , R. 1986a. Pyroxene exsolution in granulite from Fyfe Hills, Enderby Land, Antarctica: Evidence for 1000 8C metamorphic temperatures in Archean continental crust. American Mineralogist, 71, 946–954. S ANDIFORD , M. & P OWELL , R. 1986b. Deep crustal metamorphism during continental extension: modern and ancient examples. Earth and Planetary Science Letters, 79, 151–158. S ATO , K. & K AWASAKI , T. 2002. High-pressure and high-temperature experiments on the phase relations in the system of Mg-rich garnet composition (Prp75Alm25): Implication for the Fe–Mg partitioning between garnet and orthopyroxene. Polar Geoscience, 15, 66–79. S ATO , K., M IYAMOTO , T. & K AWASAKI , T. 2004. Experimental constraints of metamorphic pressure and temperature, and phase relations of a phlogopitebearing orthopyroxene granulite from Howard Hills, Napier Complex, East Antarctica. Journal of Mineralogical and Petrological Sciences, 99, 191–201. S ATO , K., M IYAMOTO , T. & K AWASAKI , T. 2006. Experimental calibration of sapphirine– spinel Fe2þ – Mg exchange thermometer: Implication for constraints on P– T condition of Howard Hills, Napier Complex, East Antarctica. Gondwana Research, 9, 398– 408. S CHALTEGGER , U., F ANNING , C. M., G UNTHER , D., M AURIN , J. C., S CHULMANN , K. & G EBAUER , D. 1999. Growth, annealing and recrystallisation of zircon and preservation of monazite in high-grade metamorphism: conventional and in-situ U– Pb isotope, cathodoluminescence and microchemical evidence. Contributions to Mineralogy and Petrology, 134, 186– 201. S HERATON , J. W. & B LACK , L. P. 1981. Geochemistry and geochronology of Proterozoic tholeiite dykes of East Antarctica: evidence for mantle metasomatism. Contributions to Mineralogy and Petrology, 78, 305–317.
JARE STUDIES IN THE NAPIER COMPLEX S HERATON , J. W. & B LACK , L. P. 1983. Geochemistry of Precambrian gneisses: relevance for the evolution of the East Antarctic Shield. Lithos, 16, 273–296. S HERATON , J. W., O FFE , L. A., T INGEY , R. J. & E LLIS , D. J. 1980. Enderby Land, Antarctica—an unusual Precambrian high-grade metamorphic terrain. Journal of the Geological Society of Australia, 27, 1– 19. S HERATON , J. W., T INGEY , R. J., B LACK , L. P., O FFE , L. A. & E LLIS , D. J. 1987. Geology of Enderby Land and Western Kemp Land, Antarctica. Bulletin Australian Bureau of Mineral Resources, 223. S HIRAISHI , K., E LLIS , D. J., F ANNING , C. M., H IROI , Y., K AGAMI , H. & M OTOYOSHI , Y. 1997. Reexamination of the metamorphic and protolith ages of the Rayner Complex, Antarctica: evidence for the Cambrian (Pan-African) regional metamorphic event. In: R ICCI , C. A. (ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antartica, Siena, 79– 88. S HIREY , S. B., H ARRIS , J. W., R ICHARDSON , S. H. ET AL . 2003. Regional patterns in the paragenesis and age of inclusions in diamond, diamond composition, and the lithospheric seismic structure of Southern Africa. Lithos, 71, 243–258. S OBOTOVICH , E. V., K AMENEV , Y. N., K OMARISTYY , A. A. & R UDNIK , N. A. 1976. The oldest rocks of Antarctica (Enderby Land). International Geology Review, 18, 371– 388. S OYAMA , Y. & I SHIZUKA , H. 2007a. Kyanite þ orthopyroxene association included in osumilite from the Mt. Riiser-Larsen area of the Napier Complex, East Antarctica; evidence for a clockwise P– T– t path. In: Abstracts of the 27th Symposium on Antarctic Geosciences, Tokyo, Japan, 56–57 [in Japanese]. S OYAMA , Y. & I SHIZUKA , H. 2007b. Prograde kyanite þ orthopyroxene association included in osumilite from the Mt. Riiser-Larsen area of the Napier Complex, Enderby Land, East Antarctica. In: Abstracts of the 4th International Symposium on Gondwana to Asia & 2007 International Association for Gondwana Research (IAGR) Annual Convention, Antarctic Geosciences, Fukuoka, Japan, 143– 144. S PEAR , F. 1993. Metamorphic Phase Equilibria and Pressure– Temperature –Time Paths. Mineralogical Society of America, Monograph. S UZUKI , S. 2000. Geochemistry and geochronology of ultra-high temperature metamorphic rocks from the Mt. Riiser-Larsen area in the Archaean Napier Complex, East Antarctica. PhD thesis, Graduate University for Advanced Studies, Tokyo. S UZUKI , S., H OKADA , T., I SHIKAWA , M. & I SHIZUKA , H. 1999. Geochemical study of granulites from Mt. Riiser-Larsen, Enderby Land, East Antarctica: Implication for protoliths of the Archaean Napier Complex. Polar Geoscience, 12, 101–125. S UZUKI , S., K AGAMI , H., I SHIZUKA , H. & H OKADA , T. 2001. Sm–Nd mineral isochron age of sapphirine– quartz gneiss from the Mt. Riiser-Larsen area in the Napier Complex, East Antarctica. Polar Geoscience, 14, 88– 98. S UZUKI , S., A RIMA , M., W ILLIAMS , I. S., S HIRAISHI , K. & K AGAMI , H. 2006. Thermal history of UHT metamorphism in the Napier Complex, East
137
Antarctica: insights from zircon, monazite, and garnet ages. Journal of Geology, 114, 65–84. S UZUKI , S., I SHIZUKA , H. & K AGAMI , H. 2008. Early to Middle Proterozoic dykes in the Mt. Riiser-Larsen area of the Napier Complex, East Antarctica: tectonic implications as deduced from geochemical studies. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 195–210. T AINOSHO , Y., K AGAMI , H., T AKAHASHI , Y., I IZUMI , S., O SANAI , Y. & T SUCHIYA , N. 1994. Preliminary result for the Sm–Nd whole-rock age of the metamorphic rocks from Mount Pardoe in the Napier Complex, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 7, 115– 121. T AINOSHO , Y., K AGAMI , H., H AMAMOTO , T. & T AKAHASHI , Y. 1997. Preliminary result for the Nd and Sr isotope characteristics of the Archaean gneisses from Mount Pardoe, Napier Complex, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 10, 92– 101. T AKAHASHI , N., K ODAIRA , S., K LEMPERER , S. L., T ATSUMI , Y., K ANEDA , Y. & S UYEHIRO , K. 2007. Crustal structure and evolution of the Mariana intra-oceanic island arc. Geology, 35, 203–206. T AKIGAMI , Y., I SHIKAWA , N. & F UNAKI , M. 1998. Preliminary 40Ar – 39Ar analyses of igneous and metamorphic rocks from the Napier Complex. Polar Geoscience, 11, 200– 207. T ATSUMI , Y. 2000. Continental crust formation by crustal delamination in subduction zones and complementary accumulation of the enriched mantle I component in the mantle. Geochemistry, Geophysics, Geosystems, 1, 2000GC000094, doi:10.1029/2000GC000094. T OYOSHIMA , T., O SANAI , Y., O WADA , M., T SUNOGAE , T., H OKADA , T. & C ROWE , A. W. 1999. Deformation of ultrahigh-temperature metamorphic rocks from Tonagh Island in the Napier complex, east Antarctica. Polar Geoscience, 12, 29– 48. T OYOSHIMA , T., Y AMAMOTO , K., O SANAI , Y., O WADA , M., T SUNOGAE , T., H OKADA , T. & C ROWE , W. A. 2001. Microstructures and deformation conditions of granulite-facies pseudotachylites from Tonagh Island, Napier Complex, East Antarctica. In: Abstracts of the 21st Symposium on Antarctic Geosciences, Tokyo, 75– 76 [in Japanese]. T OYOSHIMA , T., O SANAI , Y. & N OGI , Y. 2008. Macroscopic geological structures of the Napier and Rayner Complexes, East Antarctica. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East–West Gondwana Connection. Geological Society, London, Special Publications, 308, 139– 146. T SUNOGAE , T., O SANAI , Y., T OYOSHIMA , T., O WADA , M., H OKADA , T. & C ROWE , W. A. 1999. Metamorphic reactions and preliminary P– T estimates of ultrahigh-temperature mafic granulite from Tonagh Island in the Napier Complex, East Antarctica. Polar Geoscience, 12, 71– 86. T SUNOGAE , T., O SANAI , Y., T OYOSHIMA , T., O WADA , M., H OKADA , T. & C ROWE , W. A. 2000. Fluorine-rich
138
H. ISHIZUKA
calcic amphiboles in ultrahigh-temperature mafic granulite from Tonagh Island in the Napier Complex, East Antarctica: preliminary report. Polar Geoscience, 13, 103 –113. T SUNOGAE , T., S ANTOSH , M., O SANAI , Y. ET AL . 2001. Carbonic fluid inclusions in ultrahigh-temperature metamorphic rocks from Tonagh Island in the Archean Napier Complex, East Antarctica: A preliminary report. Polar Geoscience, 14, 25–38. T SUNOGAE , T., S ANTOSH , M., O SANAI , Y., O WADA , M., T OYOSHIMA , T. & H OKADA , T. 2002. Very highdensity carbonic fluid inclusions in sapphirine-bearing granulites from Tonagh Island in the Archean Napier Complex, East Antarctica: implications for CO2 infiltration during ultrahigh-temperature (T . 1,100 8C) metamorphism. Contributions to Mineralogy and Petrology, 143, 279– 299. T SUNOGAE , T., S ANTOSH , M., O SANAI , Y., O WADA , M., T OYOSHIMA , T., H OKADA , T. & C ROWE , W. A.
2003. Fluid inclusions in an osumilite-bearing granulite from Bunt Island in the Archean Napier Complex, East Antarctica: implication for a decompressional P– T path? Polar Geoscience, 16, 61–75. W HITEHOUSE , M. J. & P LATT , J. P. 2003. Dating high grade metamorphism—constraints from rare-earth elements in zircon and garnet. Contributions to Mineralogy and Petrology, 145, 61–74. Y OSHIMURA , Y., M OTOYOSHI , Y., G REW , E. D., M IYAMOTO , T., C ARSON , C. J. & D UNKLEY , D. J. 2000. Ultrahigh-temperature metamorphic rocks from Howard Hills in the Napier Complex, East Antarctica. Polar Geoscience, 13, 60–85. Y OSHIMURA , Y., M IYAMOTO , T., G REW , E. D., C ARSON , C. J., D UNKLEY , D. J. & M OTOYOSHI , Y. 2001. High-grade metamorphic rocks from Christmas Point in the Napier complex, East Antarctica. Polar Geoscience, 14, 53–74.
Macroscopic geological structures of the Napier and Rayner Complexes, East Antarctica T. TOYOSHIMA1, Y. OSANAI2 & Y. NOGI3 1
Graduate School of Science and Technology, Niigata University, 8050 Ikarashi-2-nocho, Niigata 950-2181, Japan, (e-mail:
[email protected]) 2
Division of Evolution of Earth Environments, Graduate School of Social and Cultural Studies, Kyushu University, Ropponmatsu 4-2-1, Fukuoka 810-8560, Japan 3
National Institute of Polar Research, Kaga 1-chome, Itabashi-ku, Tokyo 173-8515, Japan
Abstract: This paper presents a form-line map of the Napier and Rayner Complexes, East Antarctica, constructed from attitude data for foliations shown on published geological maps, and discusses the macroscopic geological structures. The form-line map shows that the two complexes consist of several, structurally distinct, units or blocks bounded by east–west-, NE–SW- and NW– SE-striking faults. The major boundary between the two complexes, as indicated on the published geological maps, is a structural discontinuity shown as a large fault on the form-line map. On the form-line map, east– west- and NE– SW-trending folds are abundant and NW– SE-trending ones occur locally in both complexes. North– south-trending folds are also abundant in the Napier Complex. Dome-and-basin fold patterns on a regional scale occur in some regions. The regional strikes, macroscopic structures, and the major boundary between the two complexes are considered to have resulted from the same later deformation episode. The form-line map and distribution map of key mineral assemblages show that the Napier Complex is not uniform and includes at least two types of metamorphic units or fragments of the Archaean crust that were formed through distinct P– T– t evolutionary processes and divided by several faults.
The Napier Complex of Enderby Land and western Kemp Land (index map in Fig. 1) is one of the several known Archaean cratonic blocks in the East Antarctic Precambrian Shield (e.g. James & Tingey 1983; Sheraton et al. 1987; Black et al. 1992). The complex is characterized by ultrahightemperature (UHT) metamorphic rocks, and is bounded on the south by younger mobile belts (Rayner Complex), like the other Archaean cratons (Fig. 1) (e.g. Sheraton et al. 1987). The structures of the Napier and Rayner Complexes have been outlined by geologists of the Bureau of Mineral Resource, Geology and Geophysics, Australia, on the basis of 1974–1977 field investigations. They have confirmed the integrity of the Napier Complex as a single highgrade Archaean craton by their assertion that the unmetamorphosed ‘Amundsen dykes’ are restricted to the complex (Sheraton et al. 1980, 1987). Sheraton et al. (1987) has reported the geological outline and structural features for the whole region of the Napier and Rayner Complexes. Toyoshima (2001) has preliminarily constructed a strike-line map of the two complexes and discussed their macroscopic geological structures. However, macroscopic geological structures and regional strikes in these two complexes remain poorly understood. The structural
geology of the boundary between the two complexes is not well known. Sheraton et al. (1987) have shown the extents of the two complexes and Napier Complex rocks overprinted by the Rayner metamorphism in their geological map and figures 4 and 16. However, geological structural relationships between the two complexes and the remetamorphosed Napier Complex rocks are not clear in the published geological maps (e.g. James & Black 1981; Ellis 1983; Sheraton et al. 1987; Harley & Hensen 1990). Harley (1998) has suggested that sapphirinebearing UHT metamorphic rocks from the northern Napier Complex represent lower pressures during the cooling stage than those from the areas south of the Amundsen Bay region, which is the highgrade part of the complex. On the basis of key mineral assemblages and reaction textures of UHT metamorphic rocks, Osanai et al. (2001b, c) and Hokada et al. (2008) have recently distinguished two metamorphic evolutionary processes; one shows isothermal decompression in part of a clockwise P–T –t path, and the other indicates isobaric cooling in part of a counterclockwise P–T– t path. They also suggested that the former path was due to the ‘true’ Napier metamorphism but the
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 139 –146. DOI: 10.1144/SP308.6 0305-8719/08/$15.00 # The Geological Society of London 2008.
Fig. 1. Form-line map of the Napier and Rayner Complexes, constructed from attitude data for foliations shown on geological maps of Sheraton et al. (1987), Ishikawa et al. (2000) and Osanai et al. (2001a). The boundary between the Napier and Rayner Complexes is from geological maps of Sheraton et al. (1987). Extensions of faults in Amundsen Bay are inferred from magnetic anomaly data of Nogi et al. (2001), and the total intensities of magnetic anomalies that are used by Nogi et al. (2001) have been included in ADMAP (Antarctic Digital Magnetic Anomaly Project) (Golynsky et al. 2001), which is shown in Figure 4. ABF, Amundsen Bay Fault; AFF, Adams Fjord Fault; CNF, Crosby Nunataks Fault; RLMSZ, Riiser-Larsen Main Shear Zone; TMF, Tula Mountains Fault.
140
T. TOYOSHIMA ET AL.
latter path shows overprinting or reworking by the Rayner metamorphism (Osanai et al. 2001b, c). Geological structural relationships between these UHT metamorphic rocks with different P–T –t paths also remain unknown. The dominant layering and lithological boundary-parallel main foliation in the Napier Complex resulted from early deformation (pre-D1 or D1 of James & Black (1981), Black & James (1983) and Sheraton et al. (1987); D1 –D2 of Toyoshima et al. (1999)). Where not significantly affected by later folding, the layering and foliation are subhorizontal, and the enveloping surfaces of the later fold generations are also subhorizontal (Sandiford & Wilson 1984). In the Rayner Complex, early deformation (Da of Sheraton et al. 1987) was also responsible for the formation of the main foliation and associated structures. These early structures have been folded and faulted locally by the later superimposed deformations. This paper presents a form-line map of the Napier and Rayner Complexes, and discusses macroscopic geological structures, regional strikes and P–T –t paths of the complexes.
Construction of form-line map Key horizons or marker beds cannot be regionally recognized in the Napier and Rayner Complexes. Therefore, a form-line map is constructed from attitude data for foliations shown on the geological maps of Sheraton et al. (1987), Ishikawa et al. (2000) and Osanai et al. (2001a). Most of data used in this paper are from Sheraton et al. (1987). Ishikawa et al. (2000) and Osanai et al. (2001a) provided data on small areas such as Mount RiiserLarsen and Tonagh Island. The main foliations on their geological maps are parallel to lithological boundaries and derived from the early deformation in each complex (e.g. Sheraton et al. 1987), and therefore we have assumed the main foliation in each complex as a structurally equivalent surface. The form-line map can indicate the forms or shapes of the regional structures involving the complexes. The contour lines at a given locality are parallel to the strike, and the spacing of the contour lines is roughly proportional to the dip angle. If contour lines are more closely spaced, the dip angle of the interval being contoured is steeper. The spacing, s, of the contour lines on the form-line map is given as s ¼ i cot u where u is a dip angle of the foliation and i ¼ 2 km is the contour interval (Fig. 1). The contour lines of the form-line map represent approximate lines of equal elevation but cannot be assigned specific values.
Before the contouring, two types of structural domains have been roughly recognized on the basis of the attitude data for foliations and lithological boundaries shown on the geological maps. One has nearly homoclinal structure, and the other shows changes in attitude of foliations. The changes indicate an abundance of folds with halfwavelengths of less than 20 km. A few examples of these structural domains and their structural features are as follows. (1) A domain around the Raggatt to southern Scott Mountains shows a north- to NWdipping homocline. (2) A domain around the northern Scott Mountains shows north-dipping foliations. (3) The southeastern part of the Nye Mountains shows a SE-dipping homocline. (4) A domain around Mount McGhee shows a NW-dipping homocline and asymmetrical minor folds. (5) A domain around Mount Stregutt to the northern Schwartz Range shows a NNE-dipping homocline and north-trending folds. (6) Changes in the orientations of foliations as well as folds are abundant in many domains around the Amundsen Bay, Tula Mountains, McDonald Ridge, the western and southern parts of Napier Mountains, Nicholas Range, southern Schwartz Range, Edward VIII Gulf, Dismal Mountains, Nye Mountains, and Mount Christensen. Fold geometry appears to be different in each domain. An interpretive contouring technique (e.g. Marshak & Mitra 1988) is used for the contour line drawing. First, contour lines are drawn in each domain. If the contour map is constructed with no faults between the domains, many linear zones of very closely spaced contours develop in the form-line map. The form-line map is then re-examined to determine if any faults may be present. Smoothed and harmonic contour lines can often be drawn by inferring faults between the structural domains and by substituting a fault for a zone of closely spaced contours (Marshak & Mitra 1988, pp. 27–34). The contour lines and fold axis are displaced along the inferred faults. The form-line map has been drawn freehand, modified by trial and error. The contour lines are drawn parallel to the lithological boundary shown on the geological maps (thick light blue lines in Fig. 1). Each of the attitude data points is assumed to represent the geological structure of an area of 20 –30 km2, if the location of one of the data points is far enough from that of other data points. We have also constructed a strike-line map of the complexes to check the inferred faults in the form-line map (Fig. 2). The strike-line map is constructed on the assumption that the thickness
MACROSCOPIC STRUCTURES OF ENDERBY LAND 141
Fig. 2. Strike-line maps of some areas in the Napier and Rayner Complexes, constructed from attitude data for foliations shown on geological maps of Sheraton et al. (1987), Ishikawa et al. (2000) and Osanai et al. (2001a).
142
T. TOYOSHIMA ET AL.
of the imaginary bed is constant. The spacing, s, of the contour lines on the strike-line map is defined by s ¼ cosec u where u is a dip angle of the foliation and the thickness of the layer is 1 km. The assumption of constant layer thickness is the general method for constructing a strike-line map. Missing, ending-off and discontinuous contours on the strike-line map correspond to the disharmonic linear zones of closely spaced contours on the form-line map drawn with no faults (Figs 2 and 3). We have inferred faults on the form-line map if the contour lines and map satisfied the following criteria (Figs 2 and 3): (1) missing, ending-off and discontinuous of fold axes; (2) disharmonic linear zones of closely spaced contours; (3) contrasting structural features in two adjacent regions and domains; (4) sudden and disharmonious changes in orientations of foliations and contours; (5) missing, ending-off and discontinuous contours on the strike-line map. The inferred faults result in drawing smoothed and harmonic contour lines (Figs 1–3).
Interpretation of form-line map The form-line map (Fig. 1) suggests structural features as follows. (1) The Napier Complex consists of several structurally distinct units or blocks separated by east– west-, NE –SW- and NW–SE-striking faults. (2) Numerous faults were recognized in the Napier Complex. We call these faults Newman Nunataks Fault, Wilkinson Peaks Fault, Napier Mountains Fault, Tippet Nunataks Fault, Northern Tula Mountains Fault, Riiser-Larsen Main Shear Zone (Ishizuka et al. 1998), Adams Fjord Fault, Crosby Nunataks Fault, Tula Mountains Fault, Beaver Glacier Fault, Amudsen Bay Fault, Napier Fault, Scott Mountains Fault, Rayner Glacier Fault, and Nye Mountains Fault, approximately from east to west (Fig. 1). A north– south-trending fault at the western part of Mount Riiser-Larsen has been referred to as the Riiser-Larsen Main Shear Zone (Ishizuka et al. 1998), but we extend the fault to the north and to the eastern Amundsen Bay where the fault trends NE– SW (Fig. 1). Nogi et al. (2001) have compiled the vector geomagnetic anomaly data obtained on board the icebreaker Shirase in Amundsen Bay. The strikes of magnetic structures in Amundsen Bay are deduced from the vector magnetic anomalies using Seama et al. (1993) and the extensions of faults in Amundsen Bay are inferred from those. These structures are
also observed in total intensity magnetic anomalies of ADMAP (Golynsky et al. 2001) that include total intensity magnetic anomaly data from the vector magnetic anomalies obtained on board the icebreaker Shirase (Fig. 4). (3) The regional strike of the Napier Complex changes from east –west in the western part to NNE– SSE to NE–SW in the central to eastern part. Where not significantly affected by later folding, the foliation dips gently north or is subhorizontal. (4) The foliation of the Rayner Complex generally strikes NE –SW and dips south. (5) NE –SW- and east– west-trending faults occur in the Rayner Complex. (6) The major boundary between the two complexes (e.g. Sheraton et al. 1987) is a structural discontinuity regarded as a fault. In the eastern part (western Kemp Land), the transition between the two complexes appears to be more gradual, as noted by Sheraton et al. (1987). (7) East–west- and NE –SW-trending folds are abundant in both complexes, but north–southtrending folds occur in some areas. (8) A dome-and-basin fold pattern on a regional scale characterizes some areas of the two complexes, as shown by James & Black (1981). (9) Zones of most closely spaced contour lines on the form-line map can be regarded as intensely folded zones or faults. Some of these structural features on a regional scale have also been implied from the recognition of structural domains and their local structural features prior to the production of the form-line map.
Discussion and conclusions Comparison of structural data and interpretation with previous studies The regional strike of the Napier Complex, shown in Fig. 1, is similar to that in figure 52 of Sheraton et al. (1987). However, the structural domains separated by the faults in the form-line map are greatly different form the sub-areas (A– L) delineated in figure 52 of Sheraton et al. (1987). The folded and homoclinal areas are recognizable in the form-line map (Fig. 1). The generalized geological map pattern shown in figure 6 of Sheraton et al. (1987) is broadly similar to the structural pattern shown in the formline map (Fig. 1). In the area from Amundsen Bay to Mount McGhee, where the distribution pattern of layered gneisses and orthogneisses is complex in figure 6 of Sheraton et al. (1987), folds and faults are abundant in the form-line map (Fig. 1). The complicated distribution pattern can be
MACROSCOPIC STRUCTURES OF ENDERBY LAND 143
Fig. 3. Form-line maps of some areas in the Napier and Rayner Complexes drawn with no faults, constructed from attitude data for foliations shown on geological maps of Sheraton et al. (1987), Ishikawa et al. (2000) and Osanai et al. (2001a).
Fig. 4. Total intensity magnetic anomaly map from ADMAP (Golynsky et al. 2001) and inferred geological structures in the Napier and Rayner Complexes. The magnetic anomaly patterns almost match the distribution of the inferred faults. Abbrevations as in Figure 1.
144
T. TOYOSHIMA ET AL.
explained by the pattern of folds and faults. In the area from the Scott Mountains to the Raggatt Mountains, where the distribution pattern of metamorphic rocks is simple on the geological map, a homoclinal structure occurs on the form-line map. Inferred faults form the boundaries between the different geological map patterns. The form-line map also indicates that the later folding and faulting gave rise to the transformation from a gently north-dipping or flat-lying structural state to a steeply inclined structural state, as shown by James & Black (1981) and Sandiford & Wilson (1984). The east– west- and NE– SWtrending folds inferred in Fig. 1 correspond to F3 of James & Black (1981), F3 of Sandiford & Wilson (1984), F3 of Sheraton et al. (1987), and B5 and B6 of Toyoshima et al. (1999), but north– south-trending folds to B4 of Toyoshima et al. (1999), on the basis of fold geometry. The east – west-, NE –SW- and NW –SE-striking faults
inferred on the form-line map can be correlated with D6 faults of Toyoshima et al. (1999), on the basis of their orientation.
Regional strikes and major structures of the Napier and Rayner Complexes The regional strike and dome-and-basin fold pattern of the Napier Complex in Fig. 1 are the products of the third deformation episode (D3 of James & Black (1981) and Sheraton et al. (1987); stage III of Toyoshima et al. (1999)). These faults and folds with near-horizontal axes have resulted from intense late-stage horizontal shortening (James & Black 1981; Toyoshima et al. 1999). The east–west- and NE–SW-trending folds, the north–south-trending folds, and the NE–SW- and east–west-striking faults also affect the south-dipping foliation of the Rayner Complex significantly (Fig. 1). The major boundary between the two complexes
Fig. 5. Distribution map of key mineral assemblages and inferred faults in the Napier Complex, complied from Sheraton et al. (1987), Ishizuka et al. (1998), Hokada et al. (1999), Osanai et al. (1999, 2001a– c) and Ishikawa et al. (2000). Abbreviations as in Figure 1.
MACROSCOPIC STRUCTURES OF ENDERBY LAND
(e.g. Sheraton et al. 1987) can be correlated with the D6 fault of Toyoshima et al. (1999), on the basis of their orientations. Therefore, the regional strike, macroscopic structures of the Rayner Complex, and the major boundary between the two complexes are considered to have largely resulted from the third deformation episode. The locations of inferred faults correspond well to the borders between high and low magnetic anomalies, which represent the boundaries of magnetic structures, on the magnetic anomaly map of the Napier and Rayner complexes shown in Fig. 4 (Golynsky et al. 2001). Figure 4 also indicates that the northern Tula Mountains Fault is likely to extend northwestward over a distance of about 60 km. The Beaver Glacier Fault and Napier Fault may also extend westward for about 100 km.
Unity of Napier Complex as a high-grade Archaean craton The highest-grade metamorphic region (Harley & Hensen 1990, p. 327, high-grade region of fig. 12.3b) of the Napier Complex appears to be surrounded by the inferred faults on the form-line map (Fig. 1). A distribution map of the key mineral assemblages (Osanai et al. 2001b, c) and the inferred faults shows that, in the Napier Complex, the faults separate the clockwise P –T– t path metamorphic units from the counterclockwise P–T –t path metamorphic units (Fig. 5). The Napier Complex is considered to be not uniform in Archaean geological history, and includes at least two types of metamorphic units or crustal fragments that were formed through distinct P–T–t evolutionary processes and bounded by the inferred faults. Our conclusion is further supported by the results obtained from the Mt. Riiser-Larsen area by Hokada et al. (2008). The western boundary of the ‘true’ Napier metamorphic rocks is not drawn as a smooth line such as shown in figure 1 of James & Black (1981) but is composed of several faults such as the Tippet Nunataks Fault, Northern Tula Mountains Fault, Riiser-Larsen Main Shear Zone, Beaver Glacier Fault, Amundsen Bay Fault and Napier Fault (Fig. 5). We wish to thank F. Storti, K. Kano and M. Ishikawa for extremely helpful reviews, and M. Satish-Kumar for editorial handling of the manuscript. Their constructive comments greatly improved the paper. H. Ishizuka, K. Shiraishi, Y. Motoyoshi, M. Owada, T. Tsunogae and T. Hokada are also thanked for their invaluable discussions and helpful advice.
References B LACK , L. P. & J AMES , P. R. 1983. Geological history of the Napier Complex of Enderby Land. In: O LIVER ,
145
R. L., J AMES , P. R. & J AGO , J. B. (eds) Antarctic Earth Science. Australian Academy of Science, Canberra, A.C.T., 11–15. B LACK , L. P., S HERATON , J. W. & K INNY , P. D. 1992. Archaean events in Antarctica. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Science. Terra, Tokyo, 1– 6. E LLIS , D. J. 1983. The Napier and Rayner Complexes of Enderby Land, Antarctica—Contrasting styles of metamorphism and tectonism. In: O LIVER , R. L., J AMES , P. R. & J AGO , J. B. (eds) Antarctic Earth Science. Australian Academy of Science, Canberra, A.C.T., 20– 24. G OLYNSKY , A., C HIAPPINI , M., D AMASKE , D. ET AL . 2001. ADMAP—Magnetic Anomaly Map of the Antarctic, 1:10 000 000 scale map. In: M ORRIS , P. & VON F RESE , R. (eds) British Antarctic Survey Miscellaneous, 10, British Antartic Survey, Cambridge. H ARLEY , S. L. 1998. On the occurrence and characterization of ultrahigh-temperature crustal metamorphism. In: T RELOAR , P. J. & O’B RIEN , P. J. (eds) What Drives Metamorphism and Metamorphic Reactions? Geological Society, London, Special Publications, 138, 81– 107. H ARLEY , S. L. & H ENSEN , B. J. 1990. Archaean and Proterozoic high-grade terranes of East Antarctica (40–808E): a case study of diversity in granulite facies metamorphism. In: A SHWORTH , J. R. & B ROWN , M. (eds) High-temperature Metamorphism and Crustal Anatexis. Unwin Hyman, London, 320– 370. H OKADA , T., O SANAI , Y., T OYOSHIMA , T., O WADA , M., T SUNOGAE , T. & C ROWE , W. A. 1999. Petrology and metamorphism of sapphirine-bearing aluminous gneisses from Tonagh Island in the Napier Complex, East Antarctica. Polar Geoscience, 12, 49– 70. H OKADA , T., M OTOYOSHI , Y., S UZUKI , S., I SHIKAWA , M. & I SHIZUKA , H. 2008. Geodynamic evolution of Mt. Riiser-Larsen, Napier Complex, East Antarctica, with reference to the UHT mineral associations and their reaction relations. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 253–282. I SHIKAWA , M., H OKADA , T., I SHIZUKA , H., M IURA , H., S UZUKI , S., T AKADA , M. & Z WARTZ , D. P. 2000. Geological map of Mount Riiser-Larsen, Enderby Land, Antarctica. Antarctic Geological Map Series, Sheet 37. National Institute of Polar Research, Tokyo. I SHIZUKA , H., I SHIKAWA , M., H OKADA , T. & S UZUKI , S. 1998. Geology of the Mt. Riiser-Larsen area of the Napier Complex, Enderby Land, East Antarctica. Polar Geoscience, 11, 154–171. J AMES , P. R. & B LACK , L. P. 1981. A review of the structural evolution and geochronology of the Archaean Napier Complex of Enderby Land, Australian Antarctic Territory. In: G LOVER , J. E. & G ROVES , D. I. (eds) Archaean Geology. Geological Society of Australia, Special Publications, 7, 71– 83. J AMES , P. R. & T INGEY , R. J. 1983. The Precambrian geological evolution of the East Antarctic metamorphic shield—a review. In: O LIVER , R. L.,
146
T. TOYOSHIMA ET AL.
J AMES , P. R. & J AGO , J. B. (eds) Antarctic Earth Science. Australian Academy of Science, Canberra, A.C.T., 5–10. M ARSHAK , S. & M ITRA , G. 1988. Basic Methods of Structural Geology. Prentice Hall, Englewood Cliffs, NJ. N OGI , Y., S EAMA , N. & I NOKUCHI , H. 2001. Magnetic anomalies in the Amundsen Bay, East Antarctica. In: Program and Abstracts for 21st Symposium on Antarctic Geosciences. National Institute of Polar Research, Tokyo, 29 [in Japanese]. O SANAI , Y., T OYOSHIMA , T., O WADA , M., T SUNOGAE , T., H OKADA , T. & C ROWE , W. A. 1999. Geology of ultrahigh-temperature metamorphic rocks from Tonagh Island in the Napier Complex, East Antarctica. Polar Geoscience, 12, 1– 28. O SANAI , Y., T OYOSHIMA , T., O WADA , M. ET AL . 2001a. Geological map of Tonagh Island, Enderby Land, Antarctica. Antarctic Geological Map Series, Sheet 38. National Institute of Polar Research, Tokyo. O SANAI , Y., T OYOSHIMA , T., O WADA , M., T SUNOGAE , T., H OKADA , T., C ROWE , W. A. & K USACHI , I. 2001b. Ultrahigh-temperature sapphirine–osumilite and sapphirine–quartz granulites from Bunt Island in the Napier Complex, East Antarctica— Reconnaissance estimation of P –T estimation. Polar Geoscience, 14, 1–24. O SANAI , Y., T OYOSHIMA , T., O WADA , M., T SUNOGAE , T., H OKADA , T., S ASANO , K. & C ROWE , W. A. 2001c. Clockwise and counter-clockwise P –T paths of UHT metamorphic rocks from the Napier Complex, East Antarctica. In: Program and Abstracts for 21st Symposium on Antarctic Geosciences.
National Institute of Polar Research, Tokyo, 20 [in Japanese]. S ANDIFORD , M. & W ILSON , C. J. L. 1984. The structural evolution of the Fyfe Hills–Khmara Bay region, Enderby Land, East Antarctica. Australian Journal of Earth Sciences, 31, 403– 426. S EAMA , N., N OGI , Y. & I SEZAKI , N. 1993. A new method for precise determination of the position and strike of magnetic boundaries using vector data of the geomagnetic anomaly field. Geophysical Journal International, 113, 155– 164. S HERATON , J. W., O FFE , L. A., T INGEY , R. J. & E LLIS , D. J. 1980. Enderby Land, Antarctica—an unusual Precambrian high-grade metamorphic terrain. Journal of the Geological Society of Australia, 27, 1– 18. S HERATON , J. W., T INGEY , R. J., B LACK , L. P., O FFE , L. A. & E LLIS , D. J. 1987. Geology of Enderby Land and Western Kemp Land, Antarctica. Australian Bureau of Mineral Resources, Geology and Geophysics, Bulletin, 223. T OYOSHIMA , T. 2001. A strike line map and geological structures of the Napier and Rayner complexes, East Antarctica. In: Program and Abstract for 21st Symposium on Antarctic Geosciences. National Institute of Polar Research, Tokyo, 77–78 [in Japanese]. T OYOSHIMA , T., O SANAI , Y., O WADA , M., T SUNOGAE , T., H OKADA , T. & C ROWE , W. A. 1999. Deformation of ultrahigh-temperature temperature metamorphic rocks from Tonagh Island in the Napier Complex, East Antarctica. Polar Geoscience, 12, 29–48.
Pre-metamorphic carbon, oxygen and strontium isotope signature of high-grade marbles from the Lu¨tzow-Holm Complex, East Antarctica: apparent age constraints of carbonate deposition M. SATISH-KUMAR1, T. MIYAMOTO2, J. HERMANN3, H. KAGAMI4, Y. OSANAI5 & Y. MOTOYOSHI6 1
Institute of Geosciences, Faculty of Science, Shizuoka University, 836 Oya, Suruga-ku, Shizuoka, 422-8529, Japan (e-mail:
[email protected]) 2
Department of Earth and Planetary Sciences, Faculty of Science, Kyushu University, Fukuoka, 812-8581, Japan 3
4
Research School of Earth Sciences, The Australian National University, Canberra A.C.T. 0200, Australia
Department of Geology, Faculty of Science, Niigata University, Ikarashi-2no-cho, Niigata, Japan 5
Division of Evolution of Earth Environment, Graduate School of Social and Cultural Studies, Kyushu University, Fukuoka, 810-8560, Japan
6
National Institute of Polar Research, Kaga, Itabashi-ku, Tokyo, 173-8515, Japan Abstract: C, O and Sr isotope geochemistry of high-grade marbles from the Lu¨tzow-Holm Complex, East Antarctica, has given clues on the depositional ages and post-depositional alterations. Dolomitic and calcitic marbles occur as thin layers with varying thickness (up to 100 m) in several outcrops in eastern Dronning Maud Land, most of which underwent postdepositional geochemical alterations. In particular, the Sr and O isotope alterations are extensive, with 87Sr/86Sr(550 Ma) ratios as high as 0.758 and d18O values as low as 25‰. These data suggest that multiple stages of fluid–rock interaction processes during diagenesis, prograde to peak and retrograde metamorphic events have altered the depositional isotopic signatures. However, some of the marble layers, exceptionally, preserve pre-metamorphic geochemical characteristics, such as low Sr isotope ratios, high d18O and d13C values, and well-equilibrated unaltered trace and rare earth element patterns. Lowest 87Sr/86Sr isotopic ratios of 0.7066 and 0.7053 with high d13C and d18O values suggest an apparent age of deposition around 730–830 Ma, although total geochemical resetting of carbonates by seawater of this age cannot be ruled out. The apparent depositional ages are consistent with carbonate deposition in the ‘Mozambique Ocean’ that separated East and West Gondwana.
Extensive metasedimentary supracrustal sequences exposed in the crustal fragments of the East Gondwana supercontinent, especially in East Antarctica, Sri Lanka, peninsular India and Madagascar, provide us with an opportunity to understand the geodynamic evolution of supercontinent assembly and breakup as well as extract key information on the depositional environments of palaeo-oceans that separated proto-continents. The closure of the Neoproterozoic ‘Mozambique Ocean’ (Hoffman 1988, 1991; Dalziel 1991; Stern 1994) is considered to be a consequence of supercontinental assembly of East Gondwana and West Gondwana during a protracted Pan-African Orogeny that spatially
extends from the Arabian–Nubian Shield to East Antarctica, through East Africa, Madagascar, Southern India and Sri Lanka (Bauer et al. 2003; Kusky & Matsah 2003). The difficulty in constraining the characteristics of the ‘Mozambique Ocean’ is mainly due to the high-grade metamorphism and tectonic reworking of the sediments during the regional Pan-African Orogeny. However, metacarbonate is a lithology that not only preserves important evidence on the metamorphic and geochemical evolution during tectonic activities (Bickle et al. 1995, 1997; Satish-Kumar et al. 1998) but also provides valuable information on the palaeo-ocean geochemistry and environment
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 147 –164. DOI: 10.1144/SP308.7 0305-8719/08/$15.00 # The Geological Society of London 2008.
148
M. SATISH-KUMAR ET AL.
of deposition (Melezhik et al. 2005). To recognize such information and to properly interpret it, a multidimensional geochemical approach has to be employed. Given the advances in microgeochemical analytical techniques, it is now possible to fingerprint and differentiate multistage geological processes accurately in space and time. In this study, we review the progress made in the field of metacarbonate geochemistry and attempt to identify pristine geochemical sedimentary signatures from high-grade marbles distributed in the Lu¨tzow-Holm Complex (LHC), East Antarctica. Lower to middle continental crust exposed in the Lu¨tzow-Holm Complex, East Antarctica (Fig. 1) is an ideal terrain to investigate crustal processes, especially for geochemical studies, because bedrocks are exposed continuously for kilometres in a single outcrop. Geological, tectonic and fluidrelated processes were the focus of several earlier studies in this region (Hiroi et al. 1983, 1986, 1987; Motoyoshi et al. 1989; Shiraishi et al. 1994, 2003; Motoyoshi & Ishikawa 1997; Satish-Kumar et al. 1998). Previous studies that documented fluid-related processes were based on mineralogy, phase petrology and fluid inclusions (Santosh & Yoshida 1998; Satish-Kumar et al. 2006a), grainscale carbon and oxygen stable isotopes (SatishKumar et al. 1998), LA-ICPMS study of trace and
rare earth elements (Satish-Kumar et al. 2006a) and Sr isotopes (Satish-Kumar et al. 2006b). Carbon, oxygen and strontium isotope studies of marbles from several outcrops exposed in the c. 400 km coastal stretch of the LHC are reviewed here. Integrating the geochemical characteristics, we discuss the influence of metamorphic fluid – rock interaction in modifying the isotopic characteristics and attempt to identify marbles that preserve pre-metamorphic signatures. We also discuss the importance of metacarbonate rocks in understanding the palaeo-ocean that separated East and West Gondwana.
Geological background and previous studies The LHC extends c. 400 km in eastern Dronning Maud Land, bounded in the NE by the Rayner Complex and in the SW by the Yamato–Belgica Complex. This region is characterized by an increase in metamorphic grade from amphibolite facies in the NE to granulite to ultrahigh-temperature metamorphic facies in the SW (Hiroi et al. 1991; Motoyoshi & Ishikawa 1997; Yoshimura et al. 2008). The dominant lithological units that crop out in this region include pelitic to psammitic gneisses,
Fig. 1. The Lu¨tzow-Holm Complex in East Antarctica, showing the localities where thick marble layers are distributed. The numbers in parenthesis are 87Sr/86Sr ratios of marbles corrected to 550 Ma.
APPARENT DEPOSITIONAL AGES OF MARBLES
149
Fig. 2. P– T fluid evolution of the Skallen region, a representative region where thick marble horizons are exposed (modified after Satish-Kumar et al. 2006a).
basic to intermediate gneisses, and subordinate calcareous and ultramafic rocks. In general, the occurrence of kyanite inclusions within garnet surrounded by sillimanite in the matrix has been reported from several outcrops in the region, indicating a regional clockwise P–T evolution (Motoyoshi et al. 1989; Hiroi et al. 1991; Yoshimura et al. 2008). Based on detailed petrological and geochemical study on scapolite-bearing rocks from the Skallen region, Satish-Kumar et al. (2006a) identified multiple fluid–rock interaction events during a protracted metamorphic evolution (Fig. 2). Detailed geochronological studies have also been carried out in this region that revealed a prolonged late Proterozoic to early Cambrian tectonothermal event, in concurrence with the widespread Pan-African events reported in the adjoining East Gondwana continental fragments (Shiraishi et al. 1994, 2003; Fraser et al. 2000; Hokada & Motoyoshi 2006). Marbles in the LHC occur as layers, up to about 100 m thick, interlayered with metapelitic and quartzofeldspathic rocks (Fig. 3a). The marble layers are often separated from the adjoining metapelitic rocks by decimetre- to metre-scale coarse-grained skarn formations, which consist of diopside, phlogopite, amphibole and spinel. Hiroi et al. (1987) carried out a detailed study on the occurrence, mineralogy and reaction textures observed in the calc-silicate rocks and marbles in the Prince Olav Coast and Soya Coast
Fig. 3. (a) Field occurrence of a thick marble layer at Skallevikshalsen. (b) Rhythmic layering of pure dolomite marble and forsterite– spinel calcite marble.
150
M. SATISH-KUMAR ET AL.
and recognized wollastonite-in and epidote-out isograds. Humite-bearing assemblages were reported from Kasumi Rock in the Prince Olav Coast (Hiroi & Kojima 1988). Recently, several new occurrences of calc-silicate rocks and marbles have been identified from the LHC (Satish-Kumar et al. 2006c). Earlier research on stable isotope characteristics of marbles from the Skallen region identified important evidence for fluid –rock interaction process. Extremely large (.20‰) oxygen isotope heterogeneities at a sub-millimetre scale were discovered in a marble layer from Skallen, which were interpreted to have resulted from grain boundary migration of aqueous fluids of meteoric origin (Satish-Kumar & Wada 1997; Satish-Kumar et al. 1998). Marbles that were not affected by fluid – rock interaction were useful in identifying the peak metamorphic temperatures using carbon isotope thermometry between calcite and graphite (Satish-Kumar & Wada 2000). In addition, recent strontium isotope studies considered a possible fluid–rock interaction event during diagenetic or early prograde metamorphism (Satish-Kumar et al. 2006a). Thus, the majority of previous studies on stable isotopes of metacarbonates rocks were focused on constraining the metamorphic evolution and less attention has been paid to understanding the pre-metamorphic features of the carbonates.
Field relations and sample descriptions Mineralogical and geochemical studies of marble samples collected from several outcrops in the LHC were carried out in this study. In particular, marbles from the layered sequence at Skallen (sampled during the 39th Japanese Antarctic Research Expedition (JARE-39) and Skallevikshalsen (sampled during JARE-46) (southwestern LHC; Fig. 1) were selected for detailed study. Major rock types in this region are orthopyroxene-bearing felsic gneiss, garnet–sillimanite gneiss, garnet–biotite gneiss and marbles (Yoshida 1977, 1978; Osanai et al. 2004). Marble layers occur conformably with thickness up to several tens of metres (Fig. 3a). Several typical high-grade marble mineral assemblages are observed, which form thin decimetre-scale rhythmic layers of pure dolomite or calcite marbles (graphite-bearing) and forsterite þ diopsideþ spinel + phlogopite + amphibole-bearing marbles (Fig. 3b). Various kinds of skarn occurrences between marble and the adjoining gneisses are also observed (Matsueda et al. 1983). Decimetre-sized scapolite- or feldspar boudins surrounded by phlogopite- or amphibole-rich reaction rims are present within the marble (Satish-Kumar
et al. 2006b, c). After careful field and hand-specimen observations we selected representative pure marble samples comprising either calcite or dolomite. Pure marbles are likely to preserve pre-metamorphic features, as the rocks might not have been affected by metamorphic processes such as fluid–rock interactions and devolatilization reactions, except for coarsening and recrystallization. To compare and contrast the effect of metamorphic processes on pre-metamorphic geochemical signatures, two representative samples from the Skallen region are studied in detail here. Sample 2305E was collected from a layer comprising an interface between a pure dolomite layer and a phlogopite– forsterite–spinel-bearing calcite-rich marble layer (Fig. 4a). Notably, the modal proportion of calcite increases drastically with a corresponding increase in the content of silicate minerals. This implies that calcite might be a product of metamorphic reactions involving dolomite and silicate phases in the precursor sediments. Furthermore, during prograde metamorphism these metamorphic (decarbonation) reactions progress with an increase in temperature and/or infiltration of aqueous fluids (Ferry 1994). Sample 2305E can therefore be considered to have been affected by metamorphic fluid–rock interaction processes. Sample 602d, a coarse-grained pure dolomite marble, was collected from a layer in the SW part of the Skallen region. Other than dolomite, the marble contains minor amounts of graphite (Satish-Kumar & Wada 2000). Silicate phases are absent in this sample and therefore this sample is considered as a potential candidate that may preserve pre-metamorphic geochemical features. Additional results from other marble horizons in the LHC, namely, Skallevikshalsen (JARE-46), Kasumi Rock (JARE-20), Kabuto Rock (JARE46) and Breidvagnipa (JARE-35), are also reported in this study. Moreover, we also take into account the existing carbon, oxygen and strontium isotopic results for the marbles from the LHC reported in the literature for discussing the regional implications (Satish-Kumar et al. 1998, 2006b; Satish-Kumar & Wada 2000). Based on evaluation of the geochemical data, we attempt to distinguish potential marbles that might preserve premetamorphic geochemical signatures that help in constraining a minimum depositional age of sedimentation.
Analytical methods Petrography of polished thin sections of marble samples was carried out using cathodoluminescence (CL) microscopy. The CL system (ELM-3R
APPARENT DEPOSITIONAL AGES OF MARBLES
151
Fig. 4. (a) Example of a stained (Alizarin red-S) marble slab (Skallen-2305E) showing pink-coloured calcite and colourless dolomite. Sampling spots with results of carbon, oxygen and strontium isotope measurements are also shown. (b) Cathodoluminescence image of marble, showing the textural relations between calcite, dolomite, forsterite and diopside from Skallen region. The green rim of diopside is formed during retrograde metamorphism. (c) Purple-coloured homogeneous CL image of coarse-grained dolomite marble from Skallen. The fractures appear bright pink.
Luminoscopew) at Shizuoka University comprises an electron gun, which can generate an electron beam with an accelerating voltage up to 30 keV and a beam current of 1 mA, fixed on the stage of a normal petrological microscope equipped
with a digital photographic attachment (Nikonw Coolpix 995). Silicate and carbonate minerals were analysed for major elements using a wavelength-dispersive electron microprobe (JEOL JXA733; Shizuoka
152
M. SATISH-KUMAR ET AL.
University). Measurement conditions were 12 nA, 15 kV accelerating voltage and a focused beam for silicate minerals, whereas a defocused beam (diameter 10 mm) was used for carbonates. Bence & Albee (1968) correction with modified a factors of Nakamura & Kushiro (1970) was performed throughout. Trace element and rare earth element (REE) analyses were performed on polished thin sections using the laser-ablation inductively coupled plasma mass spectrometry (ICP-MS) facility at the Research School of Earth Sciences (RSES) at the Australian National University (ANU), Canberra. A pulsed 193 nm ArF excimer laser with 70 mJ energy at a repetition rate of 5 Hz coupled to an Agilent 7500 quadrupole ICP-MS system was used for ablation. During the timeresolved analysis, the contamination from fractures or included phases was detected by monitoring several elements and integrating only the ‘clean’ part of the signal. A spot size of 142 mm was used for analysing coarse minerals, whereas thin rims and reaction zones were analysed using reduced spot sizes of either 82 or 54 mm. An NIST-612 glass was used as the external standard and a BCR-2G glass was used as secondary standard. Internal standards employed CaO contents measured using electron microprobe analysis (EMPA) for respective minerals. Average REE contents of several spot measurements (2–6) for each mineral textural site of representative carbonates were normalized to chondrite (McDonough & Sun 1995). Relative 1s standard deviation of analyses is generally about 5–20%. Carbon and oxygen isotopic composition were measured for carbonates (both calcite and dolomite) that were separated using a knife edge by scraping from a polished slab of marble that was previously stained with Alizarin red-S, to distinguish between calcite and dolomite. Pulverized dolomite or calcite was then placed in small stainless steel thimbles and dropped into a reaction vessel containing concentrated phosphoric acid at 60 8C in vacuum to liberate CO2 (Wada et al. 1984). The liberated CO2 gas was then purified cryogenically for analysis. Stable isotope measurements were carried out with a Finnigan MAT-250 mass spectrometer (Shizuoka University). Machine standards calibrated to NBS-20 standard yield reproducibility of 0.03‰ for d13C and 0.05‰ for d18O. The results are reported in conventional d notation related to the V-PDB standard for carbon and V-SMOW standard for oxygen, and are presented in Table 1. Carbonates were then separated for strontium isotope analysis from the same sample slabs from which sampling for stable isotopes was performed (Fig. 4a). Sample powder (3– 10 mg) was either scraped using a knife edge or drilled using a dentists
drill under a microscope. Conventional isotope dilution methods were applied to determine Rb and Sr compositions of the samples. Powdered sample fractions were dissolved with HCl– (COOH)2 mixed acid, and then passed through a Dowex 50W-X8 cation exchange resin to separate Rb and Sr. Rb and Sr compositions were determined with a Hitachi RMU5G and a JEOL JMS05RB mass spectrometer at Kyushu University (Japan). The E & A strontium standard gave values of 87 Sr/86Sr ¼ 0.7080 + 0.0001 (1s) and NBS987 gave values of 87Sr/86Sr ¼ 0.71025 + 0.00010 (1s). The contamination levels of Rb and Sr are 1 1029 g and 3 10210 g per sample. The decay constant used for 87Rb is 1.42 10211 a21 (Steiger & Ja¨ger 1977). Additional Sr isotopic ratio and concentration were analysed using a MAT 262 mass spectrometer at Niigata University (Japan), following the same procedure. Sr isotope results are summarized in Table 1.
Mineralogy and textural features Marbles from the LHC are generally coarse-grained with well-formed crystals, and show hypidiomorphic and granoblastic texture. The marbles are predominantly composed of dolomite or calcite. Other constituent minerals are forsterite, diopside, phlogopite, spinel, apatite, pargasite and tremolite. Humite-group minerals were reported from the Kasumi Rock locality (Hiroi & Kojima 1988). A minor amount of graphite is present in some of the marble horizons. All silicate mineral phases have near end-member chemical composition with only limited variations (Hiroi et al. 1987; Hiroi & Kojima 1988). CL imagery is a useful petrological indicator for identifying fluid –rock interaction processes (e.g. Yardley & Lloyd 1989; Wada et al. 1998). The majority of the marbles from the LHC display normal red to dark red CL, whereas some samples show yellow CL along fractures and grain boundaries. Yellow CL in calcite is characteristic of domains that were affected by fluid interaction and/or recrystallization (Buick & Cartwright 2003). Detailed isotopic (C, O and Sr) studies of samples that show signatures of fluid–rock interaction processes during metamorphism have been discussed by Satish-Kumar et al. (2006b). However, for comparison, new results obtained from one sample (2305E) are discussed in detail here. Coarse dolomite and calcite in this sample display purple and orange–red CL (Fig. 4b). At places, along fractures and grain boundaries, calcite displays yellow CL. Thin diopside rims around forsterite show yellowish green CL (Fig. 4b). These CL textures are characteristic of retrograde fluid–rock interaction and related chemical
Table 1. Carbon, oxygen and strontium isotope composition of marbles from Lu¨tzow-Holm Complex, East Antarctica d13C(PDB)
d18O(SMOW)
Rb (ppm)
Sr (ppm)
Calcite Calcite Dolomite Calcite Calcite
20.23 20.24 20.24 20.61 20.42
14.92 14.93 13.35 14.36 14.36
0.02 0.05 0.04 0.03 0.05
198.3 206.9 116.1 116.3 127.6
0.72534 0.72609 0.72538 0.72537 0.72497
0.00002 0.00006 0.00007 0.00008 0.00009
Dolomite Dolomite Dolomite Dolomite
1.11 0.81 1.09 0.71
17.02 17.04 16.91 16.69
0.03
170.1
0.70702
0.00006
0.03
349.2
0.70663
0.00006
Calcite Calcite
24.54 24.44
11.71 11.66
— —
— —
0.71123 0.71136
0.00002 0.00006
Calcite Calcite Calcite
22.20 22.06 22.76
14.26 14.12 13.31
0.18
109
0.70731
0.00006
0.75 0.75 0.62 0.73 1.95 1.29 1.13
19.19 18.94 19.12 19.08 19.66 19.78 19.68
0.03 —
120 74
0.72556 0.73287
0.00004 0.00001
—
33
0.73057
0.00001
— — — —
— — 255 193
0.75821 0.75739 0.70999 0.70996
0.00002 0.00003 0.00001 0.00001
Sample
87
Sr/86Sr(550 Ma)
1s
Skallen
Breidvagnipa N94-2206-31M N94-2206-60M Kasumi Rock 79020411-CC1 CC-2 CC-3 Skallevikshalsen TM99-0206-02A SK05-120-1-1-2
SK05-119-2-3-1 SK05-116-1-14
Dolomite Calcite-1 Calcite-2 Calcite-3 Calcite-4 Dolomite-1 Dolomite-2 Dolomite-3 Calcite Dolomite Calcite Dolomite
0.06 0.22
19.21 19.97
APPARENT DEPOSITIONAL AGES OF MARBLES
A97-2305E CC1 CC2 CC3 CC4 CC5 Y69-0602d CC1 CC2 CC3 Y69-0602e
153
154
M. SATISH-KUMAR ET AL.
alterations. Furthermore, exsolution of dolomite in calcite is a common feature in calcite-rich domains. In contrast, coarse monomineralic dolomite and calcite marbles display characteristic homogeneous purple and orange– red CL as shown in Figure 4c (Sample 602d), except for healed fractures, which show brighter pink CL. Exsolution of calcite in dolomite is observed; however, this feature is less prominent than in calcite-rich marbles. Given the homogeneity in CL images, monomineralic dolomite and calcite marbles are considered to have less been affected by fluid –rock interaction processes.
Carbonate mineral chemistry and metamorphic conditions Appreciable mineral chemical compositional variations are found in dolomite and calcite. In particular, the calcite-rich marbles from the southwestern LHC show pronounced exsolution textures. The Ca–Mg covariations clearly suggest that exsolution dominated compositional re-equilibration during retrograde metamorphism (Fig. 5). Dolomite displays minor variations in Fe and Mn contents, with Fe contents marginally higher than the Mn contents. The Fe and Mn contents of calcite overlap, although a slightly higher Fe than Mn content is observed in calcite. Forsterite þ spinel is the common equilibrium assemblage in marbles from the southwestern LHC. This assemblage is indicative of granulitefacies conditions. Peak metamorphic temperature estimates of 860 + 15 8C were reported from marbles in the Skallen region using calcite –graphite carbon isotope thermometry (Satish-Kumar & Wada
2000). Mizuochi et al. (2006) estimated the metamorphic temperature condition of marbles at Skallevikshalsen using calcite –dolomite solvus thermometry. The matrix calcite yielded lower temperatures corresponding to retrograde exsolution (400 –750 8C), whereas reintegrated inclusions within forsterite or spinel gave temperatures ranging from 850 to 950 8C, comparable with peak metamorphic conditions estimated from adjoining metapelitic assemblages (Yoshimura et al. 2004).
Carbonate trace element composition Carbonate trace element compositions combined with petrography and CL images provide a powerful tool to distinguish between marbles that preserve fully equilibrated calcite and dolomite, and those altered by retrograde interaction with fluids. Figure 6a shows the REE patterns of calcite and dolomite from a dolomite marble (602d) that displays well-equilibrated grains and normal red CL colours. The patterns plot in a tight range and indicate that calcite and dolomite are in REE equilibrium. Narrow deformation bands in dolomite are clearly visible in CL as pink domains (Fig. 4b). These domains are decorated by tiny fluid inclusions and appear as dusty dolomite. The REE pattern of this dusty dolomite is indistinguishable from that of the normal dolomite domains. Figure 6b displays the REE patterns of calcite and dolomite from an olivine marble (2305E) that contains carbonate grains that are partly dusty and display optical evidence of alteration, abundant trails of fluid inclusions and domains with yellow CL. Interestingly, the REE patterns of calcite still define a tight range, indicating that REE have not
Fig. 5. Mineral chemistry variations of calcite and dolomite from Skallen marbles.
APPARENT DEPOSITIONAL AGES OF MARBLES
Fig. 6. Chondrite-normalized REE diagram for carbonate minerals in (a) a ‘normal’ marble (602d: 87 Sr/86Sr ¼ 0.707) and (b) a high-Sr marble (2305E: 87 Sr/86Sr ¼ 0.725) from the Skallen region.
been significantly mobilized by potential alteration. The wider range in REE observed in dolomite is mainly produced by different proportions of exsolved calcite within the dolomite. Carbonates also contain significant amounts of fluid-mobile elements such as Sr and Pb. In sample 602d, calcite and dolomite define a tight cluster in a Sr v. Pb plot (Fig. 7). Moreover, the dusty dolomite contains only a tiny amount more Pb than the clear domains. In contrast, there is a wide scatter in carbonate Sr and Pb contents from sample 2305E. The clean calcite domains do not define a cluster and the altered calcite domains, dusty calcite with yellow CL and calcite containing abundant fluid inclusions plot in a wide field with overall higher Pb and slightly higher Sr contents, indicating abundant alteration within this sample. A similar, although less pronounced pattern is observed in dolomite. Interestingly, the domains in dolomite that contain abundant calcite exsolutions plot mainly in the field between unaltered calcite and dolomite. This indicates that the alteration responsible for the scatter in Pb and Sr occurred after the exsolution event (i.e. during retrograde metamorphism). The influx of an H2O-rich fluid phase during retrograde metamorphism, enriched
155
Fig. 7. Trace element variation (Pb v. Sr) observed in two representative marble samples (602d and 2305E) from Skallen. Sample 602d shows a tight cluster for both calcite and dolomite, whereas the data for 2305E are scattered, indicating alteration during metamorphism. Fl, fluid inclusions.
in Sr and Pb as schematically shown in Figure 2, might be responsible for the alteration.
Carbon and oxygen isotope geochemistry Data from earlier studies as well as newly obtained data on carbon and oxygen isotopic composition of marbles from various exposures in the LHC are complied in a C –O isotopic variation diagram (Fig. 8), where considerable scatter can be observed. In general, carbonate rocks deposited during late Proterozoic time have carbon and oxygen isotopic compositions of .0‰ and 25 + 5‰, respectively (Veizer et al. 1992; Halverson et al. 2005), except for narrow intervals of glaciation. In general, based on the C –O isotopic composition, marbles can be classified into three categories: marbles that possibly preserve premetamorphic isotope signatures (Type-I), marbles that follow extensive devolatilization-induced isotope trends (Type-II), and marbles that were affected by retrograde fluid infiltration events (Type-III). Type-I marbles are comparatively rare
156
M. SATISH-KUMAR ET AL.
Fig. 8. Compilation of existing carbon and oxygen isotopic composition on a d13C v. d18O variation diagram, and the results obtained in this study. The curved and straight lines were decarbonation trends of marbles by Rayleigh and batch devolatilization calculated after Baumgartner & Valley (2001). Data source for Skallen marbles are Satish-Kumar & Wada (2000) and Satish-Kumar et al. (1998, 2006b).
in the LHC. Although some of the marble horizons display carbon isotopic composition characteristic of Neoproterozoic unmetamorphosed carbonate rocks (.0‰), their oxygen isotopic values are generally lower than 20‰. Indeed, it is rather difficult to preserve pre-metamorphic oxygen isotope composition of carbonates during high-grade metamorphism, because the interaction between carbonate and externally derived aqueous fluids can easily shift oxygen isotopes even during the earliest stages of diagenesis or prograde metamorphism. Therefore, oxygen isotopes serve as a first-hand proxy for defining the extent of geochemical alteration during metamorphism. Many marbles in the LHC display coupled C–O isotopic depletion, which we group as Type-II. This trend is shown by samples that contain an appreciable amount of silicate minerals or are spatially associated with skarns, indicating the progress of decarbonation reactions (Baumgartner & Valley 2001). In addition, external fluid infiltration events along lithological contacts (see Satish-Kumar et al. 2006a) might have also altered the isotopic signatures. Type-III marbles, especially those from the Skallen region, display major oxygen isotope heterogeneity in micro-domains (Satish-Kumar et al. 1998, 2006b). Based on the extensiveness of the negative shift in oxygen isotopic composition
and the small, but consistent, shift in carbon isotopic composition, it can be considered that aqueous fluid of meteoric origin percolated the marble along fractures and grain boundaries, during the brittle regimes of retrograde metamorphism (Fig. 8). Satish-Kumar et al. (1998), based on a microscale isotope zonation study in a sample from the same horizon, suggested that dissolution–reprecipitation and further diffusion resulted in heterogeneities of the order of 20‰. In high-grade metamorphic terranes, it is not uncommon to find fluid infiltration events during retrograde metamorphism (e.g. Buick & Cartwright 1996; Buick et al. 1997; Cartwright et al. 1997). In most cases, deep circulating surface fluids alter oxygen isotopic composition of marbles to a higher extent than those fluids derived from crystallizing magmas (Buick & Cartwright 2003). The Skallen marbles were affected by deep circulating surface fluids, which penetrated through selective grain boundaries and fractures (Satish-Kumar et al. 1998). New C– O isotope results from the marbles investigated from the LHC are included in Figure 8. In particular, the d18O values (13.5 – 14.9‰) of sample 2305E are considerably lower than those of sample 602d (17‰) reported by Satish-Kumar & Wada (2000). It is therefore clear that 2305E has been affected by metamorphic
APPARENT DEPOSITIONAL AGES OF MARBLES
fluid –rock interaction processes, corroborating the evidence seen in CL imagery and trace element variations. Furthermore, d13C and d18O results from Skallevikshalsen, Breidvagnipa and Kasumi Rock marbles cluster nearer the Neoproterozoic d13C and d18O values (Fig. 8) and retain the possibility of preserving premetamorphic signatures.
Strontium isotope geochemistry Strontium isotopic composition of marbles displays remarkable variations within single marble layers as well as between layers on a regional scale in the LHC. In particular, marble layers from Skallen and Skallevikshalsen have unusually high 87 Sr/86Sr(550 Ma) ratios, whereas some layers of dolomitic marbles exhibit ‘near-normal’ Proterozoic sedimentary Sr isotopic ratios. The highest (0.7582) and lowest (0.7066) 87Sr/86Sr(550 Ma) values were observed in a dolomite marble from Skallevikshalsen and Skallen localities, respectively (Table 1). Moreover, a calcite marble from the Kabuto Rock outcrop has a 87Sr/86Sr(550 Ma) value of 0.7053 (S. Kagashima, pers. comm.). Although not in all cases, the higher 87 Sr/86Sr(550 Ma) ratios can be correlated with a decrease in d13C and d18O values (Type-II and Type-II marbles), which can be attributed to the fluid –rock interaction process as discussed by Satish-Kumar et al. (2006b).
157
For example, samples that show high Sr/86Sr(550 Ma) ratios have variations within single hand specimens (0.7250–0.7261 in sample 2305E; Fig. 4a). The d18O values are characteristically low (Fig. 4a), corresponding to the Type-II marble. However, within-sample variations in 87 Sr/86Sr(550 Ma) ratios could not be directly correlated to carbon and oxygen isotopic compositions (Fig. 4a). In contrast, monomineralic marble layers that display low strontium isotopic composition, such as 0.7070 for sample 602d, have relatively homogeneous 87Sr/86Sr(550 Ma) values and characteristically high oxygen and carbon isotope values (Fig. 8), and less scatter in trace element and REE contents (Figs 6 and 7). Therefore, marbles of Type-I have equilibrium homogeneous geochemical characteristics. In addition to the low 87Sr/86Sr(550 Ma) ratios, the Type-I marbles also have higher Sr contents (Fig. 9). However, the variation in the Sr contents could not be directly correlated with a simple mixing of sources such as those derived from the pelitic rocks, but in general, metamorphic fluid –rock interaction processes can be considered as causing the alteration of 87Sr/86Sr(550 Ma) ratios (Satish-Kumar et al. 2006b). 87
Discussion Carbon, oxygen and strontium isotopic compositions of metacarbonate rocks have been widely
Fig. 9. Sr isotopic compositions as a function of Sr contents. The scatter of Sr isotope ratios for different marble samples from the same locality should be noted. The Sr initial isotope ratio for metapelitic rocks from Skallen is after Miyamoto et al. (2008).
158
M. SATISH-KUMAR ET AL.
utilized for deducing the fluid –rock interaction processes in the continental crust (e.g. Cartwright & Valley 1991; Richards et al. 1996; Bickle et al. 1997; Buick et al. 1997; Satish-Kumar et al. 1998). Alternatively, several studies have also pointed out the potential application of carbonate or metacarbonate rocks in chemostratigraphic studies (Baker & Fallick 1988; Melezhik et al. 2001a, b, 2002a, b, 2005, 2006), because they preserve pre-metamorphic isotope signatures (Baker & Fallick 1988; Wickham & Peters 1993; Vyhnal & Chamberlain 1996; Melezhik et al. 2005, 2006). In addition to stable and radiogenic isotopes, trace element and REE geochemistry is also useful in identifying metamorphic fluid–rock interaction in marbles (Boulvais et al. 2000; Korsakov & Hermann 2006; Satish-Kumar et al. 2006b). Recent studies have also found that high-grade calcite marbles have a higher geochemical preservation potential than dolomitic marbles (Melezhik et al. 2005). Isotope geochemistry (especially Sr isotopes of seawater) is believed to be largely controlled by the supercontinent cycle, reflecting a balance between the continental input and mantle input to the oceans (Veizer et al. 1992; Condie 2000, 2003). The scarcity of unaltered or unmetamorphosed carbonate sediments in the Precambrian places limitations on our understanding of palaeo-oceans. However, metacarbonate rocks are common in the Proterozoic mobile belts. Apparently, in the absence of interaction with fluids during post-depositional events, high-grade marbles are potential candidates for understanding the palaeo-seawater chemistry and may provide insight into the determination of an age of deposition and thereby be useful in correlation studies. Conventionally, shifts from typical sedimentary signatures of carbon and oxygen isotopic composition are caused by metamorphic decarbonationrelated fluid– rock interaction or exchange with external fluids (Baumgartner & Valley 2001). Pre-metamorphic alterations, such as diagenesis or dolomitization, can also induce isotopic shifts. Decarbonation and associated fluid –rock interaction can occur through batch volatilization or Rayleigh volatilization, and in these cases the C–O isotope fractionation can be traced through modelling, by taking into account the amount of remaining carbonates and final isotopic composition (Baumgartner & Valley 2001, and references therein). However, interaction with external fluids is often complicated and difficult to trace appropriately, unless obvious geological evidence is identified. In addition to stable isotopes, Sr isotopes can also provide valuable information on the source characteristics of carbonate sediments as well as post-depositional processes. Initial Sr isotopic composition of carbonates usually preserves
chemical characteristics of seawater and, if undisturbed, indicates an apparent depositional age. However, Sr isotopic composition is highly sensitive to fluid processes during metamorphism under crustal regimes (e.g. Bickle et al. 1997; Abart et al. 2002) and is commonly used in tracing post-depositional alteration events.
Geochemical signatures of post-depositional alterations Carbonate rocks are highly vulnerable to geochemical alteration after deposition. However, in regional highgrade terranes, alterations during diagenesis and prograde metamorphism are difficult to constrain, because of the geochemical homogenization during high-temperature metamorphism and post-peak metamorphic resetting. In a recent study, Melezhik et al. (2005) evaluated the post depositional isotopic resetting based on major and trace element geochemistry, and d18O, d13C and 87Sr/86Sr ratios in Neoproterozoic dolomitic and calcitic marbles metamorphosed under amphibolite-facies conditions. Whereas both dolomitic and calcitic marbles display a high degree of preservation of premetamorphic carbon and oxygen isotopic composition, Sr isotope ratios of dolomitic marbles were reset during post-depositional alteration. Comparatively, calcitic marbles were more capable of preserving lower 87Sr/86Sr ratios than dolomitic marbles, suggesting an apparent depositional age of 700–600 Ma. The large-scale heterogeneity in Sr isotopic composition of marbles from the LHC is considered to have resulted from the post-depositional alteration of carbonates. Marbles associated with skarns have generally higher Sr contents than pure marbles. Satish-Kumar et al. (2006b) reported a 87 Sr/86Sr(550 Ma) ratio of 0.735 for Cl-rich scapolite and 0.731 for CO3-rich scapolite. Metapelitic rocks associated with the marble layer have a 87 Sr/86Sr(550 Ma) ratio of 0.764 (Miyamoto et al. 2008). These variations could not be directly correlated with simple mixing of two sources such as pelitic and carbonate rocks, although some of the samples seem to follow that trend (Fig. 9). Extensive Sr isotopic variations are observed in pure dolomitic and calcite marbles. In particular, the enrichment of Sr isotopes with concomitant decrease in Sr content is most effective during diagenesis. Further, interaction with fluids released from adjacent pelitic lithologies during prograde metamorphism can also cause large-scale shifts. A simple interaction model between carbonate rocks and diagenetic or prograde metamorphic fluids cannot explain the extensive variation observed in the Sr isotope systematics. More detailed
APPARENT DEPOSITIONAL AGES OF MARBLES
microscale analyses and assessment of Sr isotopes are necessary for understanding the postmetamorphic processes and will be addressed elsewhere. Here, we tentatively consider the large-scale variations to have occurred in an open system during post-depositional processes (diagenesis þ metamorphic fluid–rock interaction) and focus our attention on the low- 87Sr/86Sr(550 Ma) marbles. In comparison with unmetamorphosed Neoproterozic carbonate sediments (Jacobsen & Kaufman 1999; Melezhik et al. 2005), most of the LHC marbles display a general shift toward lower oxygen isotope values, higher Sr isotope ratios, and variations in trace element and REE characteristics. These characteristics are indicative of substantial geochemical resetting during postdepositional diagenesis and metamorphism. Integrating the observed geochemical features with the evidence obtained from the occurrence of Cl-rich scapolite from Skallen and Skallevikshalsen (Satish-Kumar et al. 2006b), we consider that the marbles were subjected to interaction with 87 Sr-enriched hypersaline fluids. 87Sr enrichment in the hypersaline fluids could result from the interaction with adjacent pelitic lithologies (87Sr/86Sr(550 Ma) ¼ 0.764; Miyamoto et al. 2008) in the early stages of metamorphism. Although the timing of this process cannot be determined, we tentatively assign it to an early stage of prograde metamorphism (,500 8C), coeval with the formation of Cl-rich scapolite (Satish-Kumar et al. 2006a). Furthermore, marbles (Type-III) were also locally affected by retrograde fluid –rock interaction, which resulted in extreme negative shifts in oxygen isotope value. These retrograde fluid– rock interactions were limited to grain boundaries and fractures, and caused substantial geochemical alterations (Satish-Kumar et al. 1998, 2006b).
Preservation of pristine sedimentary geochemical features Despite the extensiveness of post-depositional geochemical alteration in many of the marble layers in the LHC, exceptional marble layers (Type-I) tend to preserve pre-metamorphic geochemical features. The following geochemical signatures are considered to support this assertion. Coarse crystalline monomineralic marbles have (1) typical homogeneous CL images; (2) wellhomogenized trace element compositions and tight REE patterns; (3) similar d13C values to nonmetamorphic Neoproterozoic carbonates; (4) the highest d18O values among the marbles in the LHC, which are comparable with those of the non-metamorphic Neoproterozoic carbonates;
159
(5) low Sr isotope ratios and high Sr concentration, comparable with values for sedimentary carbonates. Early stages of chemical alteration during diagenesis and dolomitization of carbonate sediments, in the presence of external fluids, will result in an enrichment of Mn and depletion of 13C and 18O (Gao & Land 1991). In Type-I marbles we could not find any geochemical signatures that are indicative of alteration during diagenesis and/or dolomitization.
Apparent age of deposition The isotopic composition of Sr in the ocean is considered to be homogeneous, because of its long residence time. However, the 87Sr/86Sr ratio in the oceans varies with time and deviates from the mantle evolution curve in the early Proterozoic, because of the input of Sr derived from continents. The increase in 87Sr/86Sr ratio, however, is irregular during Neoproterozoic to early Palaeozoic time, and has been constrained rather precisely based on detailed studies on temporal trends in sedimentary carbonate rocks (Derry et al. 1994; Denison et al. 1998; Jacobsen & Kaufman 1999; Kuznetzow et al. 2003a, b; Melezhik et al. 2006). Although it is difficult to substantiate complete preservation of sedimentary isotopic compositions, an apparent age of deposition can be assigned from the least altered 87Sr/86Sr ratios, by comparing them with the temporal trends in 87Sr/86Sr in seawater prior to the peak metamorphism in the region (Fig. 10). In the SW LHC, from the least altered 87Sr/86Sr ratio of 0.7066, obtained from Skallen, an apparent age of c. 730 Ma could be deduced or c. 640 Ma, if the reference curve of Melezhik et al. (2006) is taken into account. Other less altered samples have 87Sr/86Sr ratios of 0.707, which indicate the possibility of a younger depositional age around 580 Ma; however, this age seems irrelevant, as discussed below. Apparently, the lowest 87Sr/86Sr ratio obtained from LHC rock is 0.7053, which might suggest carbonate deposition in the region as old as c. 830 Ma. Alternatively, these strontium isotope ratios might be preserving an event of interaction of earlier deposited carbonate sediments with seawater of those ages, which cannot be ruled out. A recent compilation of Neoproterozoic carbon isotope data of unmetamorphosed carbonate rocks in the Neoproterozoic by Halverson et al. (2005) suggests that the d13C values were positive (up to 5‰) before 670 Ma, except for short glaciation intervals. The highest d13C values of carbonates from the LHC mostly group around 0‰, with exceptional layers preserving positive values up to 2‰. The reference chemostratigraphic curve of Halverson et al. (2005) cannot be directly compared
160
M. SATISH-KUMAR ET AL.
Fig. 10. Extrapolation of initial Sr isotopic composition for a possible depositional age of carbonate rocks in the marginal sea between East and West Gondwana. The temporal trend of 87Sr/86Sr in seawater is after Melezhik et al. 2006 and references therein.
with the carbon isotope values of the marbles in the present study, as most of the marbles contain graphite. The presence of graphite indicates that the original carbon isotope values of calcite were re-equilibrated with organic carbon having low d13C during metamorphism. Alternatively, the d13C values of carbonates now preserve the minimum values and do not contradict the depositional ages estimated using Sr isotope ratios. With the limitations described above, we tentatively assign the age spectrum of deposition of carbonate rocks in the LHC to be between 730 and 830 Ma. Further detailed geochemical and isotopic study of carbonate rocks is necessary to confirm and refine this age spectrum. The age of sedimentation obtained from the carbonate rocks is in broad agreement with the existing geochronological and tectonic framework of the LHC. Detailed U– Pb sensitive high-resolution ion microprobe (SHRIMP) studies on metamorphic gneisses in this region have suggested that the age of peak metamorphism is c. 530 Ma (Shiraishi et al. 1994, 2003), although earlier ages of around 650 –580 Ma have also been reported recently (Hokada & Motoyoshi 2006). The early metamorphic ages suggest that the deposition of carbonate rocks could not have occurred as late as 580 Ma, but instead might represent an alteration event during metamorphism. The age of sedimentation suggested in this study is also supported by the youngest reported inherited ages of magmatic zircons in metapelitic rocks in the LHC region. Shiraishi et al. (2003) reported
c. 1000 Ma age zircon cores from several localities in the LHC. They considered the protolith of the pelitic gneisses as detritus deposited at the margin of a pre-Grenvillian craton, which consists of various continental components of ages older than 1000 Ma. Moreover, Shiraishi et al. (2008) reported a youngest TDM age of 0.87 Ga and a major mode at 1.0–1.25 Ga for samples from the LHC. This provides an upper age limit for sedimentation in the region. The results of our study supplement this assertion that the deposition of sediments in the ‘Mozambique Ocean’ between East and West Gondwana might have occurred between 730 and 830 Ma. Carbonate rocks preserve important evidence on the marginal sea conditions of supercontinent assembly. In particular, the Sr isotope record of carbonate can give insights into the palaeo-ocean chemistry. A sharp decrease in Sr isotope ratio during the early Neoproterozoic (c. 850 Ma) is evident in many recent studies of unmetamorphosed carbonate sediments (see Melezhik et al. 2006, and references therein). Low Sr isotope ratios are indicative of enhanced mantle input of Sr to the seawater, which potentially relates to the breakup of supercontinents. In contrast, a steady increase in 87 Sr/86Sr ratio in the seawater might result from the supply of Sr during extensive continental erosion, following continental uplift during collision. For example, the steep increase in the Sr isotope ratio between 600 and 500 Ma corresponds to possible collision between East and West
APPARENT DEPOSITIONAL AGES OF MARBLES
Gondwana to form the Gondwana supercontinent (Condie 2000). Thus, strontium isotopes serve as an indirect proxy in understanding the supercontinent cycle. Thick metasedimentary sequences exposed in East Antarctica include metacarbonate rocks supposed to have been deposited in the ‘Mozambique Ocean’ that separated East and West Gondwana (Hoffman 1988, 1991; Dalziel 1991; Stern 1994). Similar to the results on high-grade marbles that we have presented here, geochemical studies may provide insight into the seawater geochemistry and may serve as valuable indicators for supercontinent evolution.
Conclusions (1) Metacarbonate rocks can provide important clues for understanding the palaeo-oceans that existed between proto-continents. (2) Geochemical studies of metacarbonate rocks from the Lu¨tzow-Holm Complex, East Antarctica, suggest that most of the metacarbonate rocks have been altered during protracted postdepositional tectonic events. (3) Most of the marbles display extreme variations in carbon, oxygen and strontium isotopic composition, suggesting an active role of fluids in the Lu¨tzow-Holm Complex continental crust. (4) Exceptional metacarbonates, which preserve possible near-depositional geochemical features, are found in the Lu¨tzow-Holm Complex. These metacarbonates constrain a possible apparent depositional age between 730 and 830 Ma. (5) There is need for more information on metacarbonates from adjacent Gondwana terranes to understand the Neoproterozoic evolution of the ‘Mozambique Ocean’ that separated East and West Gondwana. M.S.-K. acknowledges a grant (No. 18740319) from the Ministry of Education, Culture, Sports, Science and Technology, Japan. We thank the geological field survey group and the crew members of the Japanese Antarctic Research Expedition for their support during the field expedition. T. Tsuchiya (Tohoku University) and S. Kagashima (Yamagata University) are thanked for sparing marble samples from Breidvagnipa and Kabuto Rock, respectively. We thank V. Melezhik and an anonymous reviewer for constructive comments, and K. Shiraishi for editorial handling of our manuscript.
References A BART , R., B ADERTSCHER , N., B URKHARD , M. & P OVODEN , E. 2002. Oxygen, carbon and strontium isotope systematics in two profiles across the Glarus
161
thrust: implications for fluid flow. Contributions to Mineralogy and Petrology, 143, 192 –208. B AKER , J. & F ALLICK , A. E. 1988. Evidence from Lewisian limestone for isotopically heavy carbon in two-thousand million year-old sea water. Nature, 337, 352–354. B AUER , W., T HOMAS , R. J. & J ACOBS , J. 2003. Proterozoic– Cambrian history of Dronning Maud Land in the context of Gondwana assembly. In: Y OSHIDA , M., W INDLEY , B. F. & D ASGUPTA , S. (eds) Proterozoic East Gondwana, Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 247– 270. B AUMGARTNER , L. P. & V ALLEY , J. W. 2001. Stable isotope transport and contact metamorphic fluid flow. In: V ALLEY , J. W. & C OLE , D. R. (eds) Stable Isotope Geochemistry. Mineralogical Society of America, Reviews in Mineralogy, 43, 415– 468. B ENCE , A. E. & A LBEE , A. L. 1968. Empirical correction factors for the electron microanalysis of silicates and oxides. Journal of Geology, 76, 382– 403. B ICKLE , M. J., C HAPMAN , H. J., W ICKHAM , S. M. & T AYLOR , H. P. 1995. Strontium and oxygen isotope profiles across marble– silicate contacts, Lizzies Basin, East Humboldt Range, Nevada: constraints on metamorphic fluid flow. Contributions to Mineralogy and Petrology, 121, 400– 413. B ICKLE , M. J., C HAPMAN , H. J., F ERRY , J. M., R UMBLE , D. I. & F ALLICK , A. E. 1997. Fluid flow and diffusion in the Waterville limestone, south– central Maine: Constraints from strontium, oxygen and carbon isotope profiles. Journal of Petrology, 38, 1489– 1512. B OULVAIS , P., F OURCADE , S., M OINE , B., G RUAU , G. & C UNEY , M. 2000. Rare-earth element distribution in granulite-facies marbles: a witness of fluid-rock interaction. Lithos, 53, 117– 126. B UICK , I. S. & C ARTWRIGHT , I. 1996. Fluid–rock interaction during low-pressure polymetamorphism of Reynolds Range Group, central Australia. Journal of Petrology, 37, 1097– 1124. B UICK , I. S. & C ARTWRIGHT , I. 2003. Fracturecontrolled fluid flow and metasomatism in the contact aureole of the Marulan Batholith (New South Wales, Australia). Contributions to Mineralogy and Petrology, 143, 733–749. B UICK , I. S., C ARTWRIGHT , I. & W ILLIAMS , I. S. 1997. High-temperature retrogression of granulite-facies marbles from Reynolds Range Group, central Australia: Phase equilibria, isotopic resetting and fluid fluxes. Journal of Petrology, 38, 877 –910. C ARTWRIGHT , I. & V ALLEY , J. W. 1991. Steep oxygen-isotope gradients at marble–metagranite contact in the northwest Adirondack mountains, New York, USA: product of fluid hosted diffusion. Earth and Planetary Science Letters, 107, 148–163. C ARTWRIGHT , I., B UICK , I. S. & M AAS , R. 1997. Fluid flow in marbles at Jervois, central Australia: oxygen isotope disequilibrium and zoning produced by decoupling of mineralogical and isotopic resetting. Contributions to Mineralogy and Petrology, 128, 335–351. C ONDIE , K. C. 2000. Episodic continental growth models: afterthoughts and extensions. Tectonophysics, 322, 153– 162.
162
M. SATISH-KUMAR ET AL.
C ONDIE , K. C. 2003. Supercontinents, superplumes and continental growth: the Neoproterozoic record. In: Y OSHIDA , M., W INDLEY , B. F. & D ASGUPTA , S. (eds) Proterozoic East Gondwana, Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 1 –21. D ALZIEL , I. W. D. 1991. Pacific margins of Laurentia and east Antarctica–Australia as a conjugate rift pair: evidence and implications for an Eocambrian Supercontinent. Geology, 19, 598– 601. D ENISON , R. E., K OEPNICK , R. B., B URKE , W. H. & H ETHERINGTON , E. A. 1998. Construction of the Cambrian and Ordovician seawater 87Sr/86Sr curve. Chemical Geology, 152, 325– 340. D ERRY , L. A., B RASIER , M. D., C ORFIELD , R. M., R ZANOV , A. Y. & Z HURAVLEV , A. Y. 1994. Sr and C isotopes in lower Cambrian carbonates from Siberian craton: a paleoenvironmental record during the ‘Cambrian explosion’. Earth and Planetary Science Letters, 128, 671 –681. F ERRY , J. M. 1994. Overview of the petrological record of fluid flow during regional metamorphism in northern New England. American Journal of Science, 294, 905– 988. F RASER , G., M C D OUGALL , I., E LLIS , D. J. & W ILLIAMS , I. S. 2000. Timing and rate of isothermal decompression in Pan-African granulites from Rundvagshetta, East Antarctica. Journal of Metamorphic Geology, 18, 441–454. G AO , G. & L AND , L. S. 1991. Geochemistry of Cambro-Ordovician Arbuckle limestone, Oklahoma: implication for diagenetic d18O alteration and d13C and 87Sr/86Sr variations. Geochimica et Cosmochimica Acta, 55, 2911– 2920. H ALVERSON , G. P., H OFFMAN , P. F., S CHRAG , D. P., M ALOOF , A. C. & R ICE , A. H. N. 2005. Toward a Neoproterozoic composite carbon-isotope record. Geological Society of America Bulletin, 117, 1181–1207. H IROI , Y. & K OJIMA , H. 1988. Petrology of dolomitic marbles from Kasumi Rock, Prince Olav Coast, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 2, 96–109. H IROI , Y., S HIRAISHI , K., Y ANAI , K. & K IZAKI , K. 1983. Aluminum silicates in the Prince Olav and Soya Coasts, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 28, 115– 131. H IROI , Y., S HIRAISHI , K., M OTOYOSHI , Y., K ANISAWA , S., Y ANAI , K. & K IZAKI , K. 1986. Mode of occurrence, bulk chemical compositions, and mineral textures of ultramafic rocks in the Lu¨tzow-Holm complex, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 43, 62– 84. H IROI , Y., S HIRAISHI , K., M OTOYOSHI , Y. & K ATSUSHIMA , T. 1987. Progressive metamorphism of calc-silicate rocks from the Prince Olav and Soya Coasts, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 1, 73– 97. H IROI , Y., S HIRAISHI , K. & M OTOYOSHI , Y. 1991. Late Proterozoic paired metamorphic complexes in East Antarctica, with special reference to the tectonic significance of ultramafic rocks. In: T HOMSON , M. R. A., C RAME , J. A. & T HOMSON , J. W. (eds)
Geological Evolution of Antarctica. Cambridge University Press, Cambridge, 83–87. H OFFMAN , P. F. 1988. United plates of America, the birth of a craton: early Proterozoic assembly and growth of Laurentia. Annual Review of Earth and Planetary Sciences, 16, 543–603. H OFFMAN , P. F. 1991. Did the breakout of Laurentia turn Gondwana inside out? Science, 252, 1409– 1412. H OKADA , T. & M OTOYOSHI , Y. 2006. Electron microprobe technique for U– Th–Pb and REE chemistry of monazite, and its implications for pre-, peak- and post-metamorphic events of the Lu¨tzow-Holm Complex and the Napier Complex, East Antarctica. Polar Geoscience, 19, 118– 151. J ACOBSEN , S. B. & K AUFMAN , A. J. 1999. The Sr, C and O isotopic evolution of Neoproterozoic seawater. Chemical Geology, 161, 37– 57. K ORSAKOV , A. V. & H ERMANN , J. 2006. Silicate and carbonate melt inclusions associated with diamonds in deeply subducted carbonate rocks. Earth and Planetary Science Letters, 241, 104–118. K USKY , T. M. & M ATSAH , M. I. 2003. Neoproterozoic dextral faulting on the Najd Fault System, Saudi Arabia, preceded sinistral faulting and escape tectonics related to closure of the Mozambique Ocean. In: Y OSHIDA , M., W INDLEY , B. F. & D ASGUPTA , S. (eds) Proterozoic East Gondwana, Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 327–362. K UZNETZOW , A. B., O VCHINNIKOVA , G. V., G OROKHOV , I. M., K AUROVA , O. K., K RUPENIN , M. T. & M ASLOV , A. V. 2003a. Sr isotope signature and Pb– Pb age of the Baikal formation limestones, the lower Riphean type section, the South Urals. Transactions of the Russian Academy of Sciences, 391, 819– 822. K UZNETZOW , A. B., S EMIKHATOV , M. A., G OROKHOV , I. M., M ELNIKOV , N. N., K ONSTANTINOVA , G. V. & K UTYAVIN , E. P. 2003b. Sr isotope composition in carbonates of the Karatau Group, southern Urals, and standard curve of 87Sr/86Sr variations in the Late Riphean ocean. Stratigraphy and Geological Correlation, 11, 415–449. M ATSUEDA , H., M OTOYOSHI , Y. & M ATSUMOTO , Y. 1983. Mg– Al skarn of the Skallevikshalsen on the east coast of Lu¨tzow-Holm Bay, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 28, 166–182. M C D ONOUGH , W. F. & S UN , S. S. 1995. The composition of the Earth. Chemical Geology, 120, 223– 253. M ELEZHIK , V. A., G OROKHOV , I. M., K USNETZOV , A. B. & F ALLICK , A. E. 2001a. Chemostratigraphy of the Neoproterozoic carbonates: implications for ‘blind experiments’. Terra Nova, 13, 1–11. M ELEZHIK , V. A., G OROKHOV , I. M., F ALLICK , A. E. & G JELLE , S. 2001b. Strontium and carbon isotope geochemistry applied to dating of carbonate sedimentation: an example from high grade rocks of Norwegian Caledonides. Precambrian Research, 108, 267– 292. M ELEZHIK , V. A., R OBERTS , D., G OROKHOV , I. M., F ALLICK , A. E., Z WAAN , K. B., K USNETZOV , A. B. & P OKROVSKY , A. B. 2002a. Isotopic evidence for a complex Neoproterozoic to Silurian rock assemblage
APPARENT DEPOSITIONAL AGES OF MARBLES in the North– Central Norwegian Caledonides. Precambrian Research, 114, 55– 86. M ELEZHIK , V. A., G OROKHOV , I. M., F ALLICK , A. E., R OBERTS , D., K USNETZOV , A. B., Z WAAN , B. K. & P OKROVSKY , B. G. 2002b. Isotopic stratigraphy suggests Neoproterozoic ages and Laurentian ancestry for high-grade marbles from the North– Central Norwegian Caledonides. Geological Magazine, 139, 375–393. M ELEZHIK , V. A., R OBERTS , D., F ALLICK , A. E., G OROKHOV , I. M. & K USNETZOV , A. B. 2005. Geochemical preservation potential of high-grade marble versus dolomite marble: implication for isotope chemostratigraphy. Chemical Geology, 216, 203–224. M ELEZHIK , V. A., K USNETZOV , A. B., F ALLICK , A. E., S MITH , R. A., G OROKHOV , I. M., J AMAL , D. & C ATUANE , F. 2006. Depositional environments and an apparent age for the Geci meta-limestones: Constraints on the geological history of northern Mozambique. Precambrian Research, 148, 19–31. M IYAMOTO , T., S ATISH -K UMAR , M., D UNKLEY , D. J., O SANAI , Y., Y OSHIMURA , Y., M OTOYOSHI , Y. & C ARSON , C. J. 2008. Post-peak (,530 Ma) thermal history of Lu¨tzow-Holm Complex, East Antarctica, based on Rb– Sr and Sm–Nd mineral chronology. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 165– 181. M IZUOCHI , H., S ATISH -K UMAR , M., T SUNOGAE , T. & M OTOYOSHI , Y. 2006. Constraints on peak metamorphic temperature and halogen content from marbles in Skallevikshalsen, East Antarctica. Proceedings of the 26th Symposium on Polar Geosciences, National Institute of Polar Research, Tokyo, 13– 15. M OTOYOSHI , Y. & I SHIKAWA , M. 1997. Metamorphic and structural evolution of granulites from Rundvagshetta, Lu¨tzow-Holm Bay, East Antarctica. In: R ICCI , C. A. (eds) The Antarctic Region: Geological Evolution and Processes. Terra Antarctic, Sienna, 65– 72. M OTOYOSHI , Y., M ATSUBARA , S. & M ATSUEDA , H. 1989. P –T evolution of the granulite-facies rocks of the Lu¨tzow-Holm Bay region, East Antarctica. In: D ALY , J. S. (ed.) Evolution of Metamorphic Belts. Geological Society, London, Special Publications, 43, 325–329. N AKAMURA , Y. & K USHIRO , I. 1970. Equilibrium relations of hypersthene, pigeonite and augite in crystallizing magmas: microprobe study of a pigeonite andesite from Weiselberg, Germany. American Mineralogist, 55, 1999–2015. O SANAI , Y., T OYOSHIMA , T., O WADA , M. ET AL . 2004. Geological map of Skallen, East Antarctica (revised version). Antarctic Geologic Map Series, Sheet 39, National Institute of Polar Research, Tokyo. R ICHARDS , I. J., L ABOTKA , T. C. & G REGORY , R. T. 1996. Contrasting stable isotope behavior between calcite and dolomite marbles, Lone Mountain, Nevada. Contributions to Mineralogy and Petrology, 123, 202– 221.
163
S ANTOSH , M. & Y OSHIDA , M. 1998. A petrologic and fluid inclusion study of charnockites from the Lu¨tzow-Holm Bay region, East Antarctica: evidence for fluid-rich metamorphism in the lower crust. Lithos, 29, 107– 126. S ATISH -K UMAR , M. & W ADA , H. 1997. Meteoric water infiltration in Skallen marbles, East Antarctica: oxygen isotopic evidence. Proceedings of NIPR Symposium on Antarctic Geosciences, 10, 111–119. S ATISH -K UMAR , M. & W ADA , H. 2000. Carbon isotopic equilibrium between calcite and graphite in Skallen Marbles, East Antarctica: evidence for preservation of peak metamorphic temperatures. Chemical Geology, 166, 172–183. S ATISH -K UMAR , M., Y OSHIDA , M., W ADA , H., N IITSUMA , N. & S ANTOSH , M. 1998. Fluid flow along microfractures in calcite from a marble from East Antarctica: evidence from gigantic (21‰) oxygen isotopic zonation. Geology, 26, 251–254. S ATISH -K UMAR , M., H ERMANN , J., T SUNOGAE , T. & O SANAI , Y. 2006a. Carbonation of Cl-rich scapolite boudins in Skallen, East Antarctica: evidence for changing fluid condition in the continental crust. Journal of Metamorphic Geology, 24, 241– 261. S ATISH -K UMAR , M., H ERMANN , J., M IYAMOTO , T. & O SANAI , Y. 2006b. Fingerprinting fluid processes in the continental crust: an integrated approach using grain-scale Sr, C, O isotopes and REE geochemistry. Geochimica et Cosmochimica Acta, 70, A557. S ATISH -K UMAR , M., M OTOYOSHI , Y., S UDA , Y., H IROI , Y. & K AGASHIMA , S. 2006c. Calc-silicate rocks and marbles from Lu¨tzow-Holm Complex, East Antarctica, with special reference to the mineralogy and geochemical characteristics of calc-silicate mega-boudins from Rundva˚gshetta. Polar Geoscience, 19, 37–61. S HIRAISHI , K., E LLIS , D. J., H IROI , Y., F ANNING , C. M., M OTOYOSHI , Y. & N AKAI , Y. 1994. Cambrian orogenic belt in East Antarctica and Sri Lanka: implications for Gondwana assembly. Journal of Geology, 102, 47– 65. S HIRAISHI , K., H OKADA , T., F ANNING , C. M., M ISAWA , K. & M OTOYOSHI , Y. 2003. Timing of thermal events in eastern Dronning Maud Land, East Antarctica. Polar Geoscience, 16, 76– 99. S HIRAISHI , K., D UNKLEY , D. J., H OKADA , T., F ANNING , C. M., K AGAMI , H. & H AMAMOTO , T. 2008. Geochronological constraints on the Late Proterozoic to Cambrian crustal evolution of eastern Dronning Maud Land, East Antarctica: a synthesis of SHRIMP U– Pb age and Nd model age data. Geodynamic Evolution of East Antarctica: a Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 21– 68. S TEIGER , R. H. & J A¨ GER , E. 1977. Subcommission on geochronology: convention on the use of decay constants in geo- and cosmo-chronology. Earth and Planetary Science Letters, 36, 359–362. S TERN , R. J. 1994. Arc assembly and continental collision in the Neoproterozoic East African orogen: implications for the consolidation of Gondwanaland. Annual Review of Earth and Planetary Sciences, 22, 319– 351.
164
M. SATISH-KUMAR ET AL.
V EIZER , J., P LUMB , K. A., C LAYTON , R. N., H INTON , R. W. & G ROTZINGER , J. P. 1992. Geochemistry of Precambrian carbonates: V. Late Proterozoic seawater. Geochimica et Cosmochimica Acta, 56, 2487– 2501. V YHNAL , C. R. & C HAMBERLAIN , C. P. 1996. Preservation of early isotopic signatures during prograde metamorphism, Eastern Vermont. American Journal of Science, 296, 294–419. W ADA , H., F UJII , N. & N IITSUMA , N. 1984. Analytical method of stable isotope for ultra-small amounts of carbon dioxide with MAT-250 mass spectrometer. Geoscience Reports of Shizuoka University, 10, 103– 112 [in Japanese with English abstract]. W ADA , H., A NDO , T. & S UZUKI , M. 1998. The role of the grain boundary at chemical and isotopic fronts in marble during contact metamorphism. Contributions to Mineralogy and Petrology, 132, 309–320. W ICKHAM , S. M. & P ETERS , M. T. 1993. High d13C Neoproterozoic carbonate rocks in western North America. Geology, 21, 165–168. Y ARDLEY , B. W. D. & L LOYD , G. E. 1989. An application of cathodoluminescence microscopy to the study of textures and reactions in high-grade marbles from Connemara, Ireland. Geological Magazine, 126, 333–337.
Y OSHIDA , M. 1977. Geology of the Skallen region, Lu¨tzow-Holmbukta, East Antarctica. Memoirs of National Institute of Polar Research Series C (Earth Science), 11, 38. Y OSHIDA , M. 1978. Tectonics and petrology of charnockites around Lu¨tzow-Holmbukta, East Antarctica. Journal of Geosciences, Osaka City University, 21, 65–152. Y OSHIMURA , Y., M OTOYOSHI , Y., M IYAMOTO , T., G REW , E. S., C ARSON , C. J. & D UNKLEY , D. J. 2004. High-grade metamorphic rocks from Skallevikshalsen in the Lu¨tzow-Holm Complex, East Antarctica: Metamorphic conditions and possibility of partial melting. Polar Geoscience, 17, 57–88. Y OSHIMURA , Y., M OTOYOSHI , Y. & M IYAMOTO , T. 2008. Sapphirine þ quartz association in garnet: implication for ultrahigh-temperature metamorphism in Rundva˚gshetta, Lu¨tzow-Holm Complex, East Antarctica. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctic: A Key to the East–West Gondwana Connection. Geological Society, London, Special Publications, 308, 377–390.
Post-peak (<530 Ma) thermal history of Lu¨tzow-Holm Complex, East Antarctica, based on Rb– Sr and Sm –Nd mineral chronology TOMOHARU MIYAMOTO1, M. SATISH-KUMAR2, DANIEL J. DUNKLEY3, YASUHITO OSANAI4, YASUTAKA YOSHIMURA5, YOICHI MOTOYOSHI3 & CHRISTOPHER J. CARSON6 1
Department of Earth and Planetary Sciences, Kyushu University, Hakozaki 6-10-1, Fukuoka 812-8581, Japan (e-mail:
[email protected]) 2
Institute of Geosciences, Faculty of Science, Shizuoka University, 836 Oya, Suruga-ku, Shizuoka, 422-8529, Japan
3
National Institute of Polar Research, 9-10, Kaga 1-chome, Itabashi-ku, Tokyo 173-8515, Japan
4
Division of Evolution of Earth Environment, Graduate School of Social and Cultural Studies, Kyushu University, 4-2-1 Ropponmatsu, Chuo-ku, Fukuoka, 810-8560, Japan
5
Department of Geology, Kochi University, 5-1, 2-chome, Akebono-cho, Kochi 780-8520, Japan 6
GA Geochronology Laboratory, Geoscience Australia, PO Box 378, Canberra, A.C.T. 2601, Australia
Abstract: Rb–Sr and Sm–Nd mineral dating of metamorphic rocks from Skallen, Skallevikshalsen and Rundva˚gshetta, in the southwestern part of the Lu¨tzow-Holm Complex, Dronning Maud Land, assists in constructing a thermal history after peak metamorphism. The results fall into two groups: (1) a record of regional cooling after peak metamorphism (524–488 Ma); (2) local resetting 50– 80 Ma after peak metamorphism (474–446 Ma). This grouping is consistently observed in published ages from various localities in the Lu¨tzow-Holm Complex. A Sm– Nd age of 524 Ma is indistinguishable from published zircon and monazite ages. Ages of 511 and 488 Ma are related to cooling after peak metamorphism. The younger age group overlaps with ages of post-metamorphic magmatism and related hydrothermal activity reported from localities throughout East Antarctica. This intracontinental, post-orogenic igneous activity continued after the tectonic assembly of Gondwana.
Geochronological studies are central in understanding the tectonic evolution of continental crust. Recent advances in microanalytical techniques, especially with the application of the sensitive highresolution ion microprobe (SHRIMP) and electron microprobe chemical dating, have revolutionized our understanding of tectonic processes (e.g. Shiraishi et al. 1994, 2003; Asami et al. 1997; Hokada & Motoyoshi 2006, for the Lu¨tzow-Holm Complex). Precise ages obtained from zircon, monazite and other U/Th-bearing minerals, supported by petrographic and geochemical information, are commonly interpreted as representing the timing of magmatism, peak metamorphism, or beginning of retrograde events in the evolution of an orogen. However, P –T– t evolution subsequent to peak metamorphism, especially in high-grade terranes, is also important for understanding the evolution of continental crust, as most such terranes are affected by igneous and hydrothermal activity
during exhumation that can significantly modify the crust. Characterizing the post-peak metamorphic history of a terrane requires careful application of mineral or rock isotope systems that record such events and often can be addressed by major rock-forming minerals, rather than accessory phases that are difficult to relate to petrography. Parameters that are critical in understanding retrograde histories include mineral-element diffusion closure temperatures, the influence of fluids and cooling rates. For example, Fraser & McDougall (1995) proposed a post-peak history of the ultrahightemperature (UHT) Rundva˚gshetta locality in the Lu¨tzow-Holm Complex (LHC) of East Antarctica, which involved rapid cooling after decompression. Furthermore, Fraser et al. (2000) elegantly demonstrated the application of 40Ar/39Ar ages of various minerals in defining a cooling history for the LHC. These studies signify the importance of obtaining results from multiple isotope systems in various
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 165 –181. DOI: 10.1144/SP308.8 0305-8719/08/$15.00 # The Geological Society of London 2008.
166
T. MIYAMOTO ET AL.
minerals or rocks in a single locality. In this study, we attempt to utilize conventional Sm –Nd and Rb–Sr mineral isochron dating to unravel the postpeak metamorphic history of a suite of rocks from the southwestern LHC. The results are discussed in a tectonic context, to enhance understanding of the geodynamic evolution of the LHC and adjoining areas.
Regional geology and previous geochronology The Lu¨tzow-Holm Complex (LHC) of Dronning Maud Land, East Antarctica, situated to the west of the Rayner Complex and to the east of the Yamato– Belgica Complex, is a high-grade metamorphic terrane within the East Antarctic Shield, where small coastal exposures are found between 458 and 378E (Fig. 1). Japanese Antarctic Research Expeditions (JARE) have carried out detailed surveys of its geology and tectonics, beginning in 1956, and have identified various kinds of metamorphic rocks, granites, pegmatites and ultrapotassic rocks that intruded during different stages of
tectonism, especially during and after peak metamorphism. The metamorphic grade of the LHC increases from upper amphibolite facies in the NE to granulite facies in the SW of the complex (Fig. 1), with a thermal maximum at Rundva˚gshetta (Hiroi et al. 1991). Recent studies have suggested a possibility of ultrahigh-temperature (UHT) metamorphism at Rundva˚gshetta (Motoyoshi et al. 2006; Yoshimura et al. 2008). Petrological studies showed that the LHC experienced metamorphism with a clockwise P– T path and isothermal decompression after the thermal peak (e.g. Motoyoshi et al. 1989; Hiroi et al. 1991). Late to post-tectonic magmatism is prominent in the LHC, including ultrapotassic intrusions exposed at Innhovde in the western LHC and neighbouring terranes (Sør Rondane, Yamato –Belgica, and Napier Complexes), which indicate a common tectonic association within Gondwana during the Ordovician (Miyamoto et al. 2000). The objectives of geochronological studies in the LHC (Table 1) have been twofold: (1) correlation with metamorphic complexes in adjacent continental fragments of Gondwana, particularly Sri Lanka and South India; (2) construction of a
Fig. 1. Regional map showing published and new Rb– Sr and Sm– Nd mineral isochron ages in the Lu¨tzow-Holm Complex, East Antarctica. Definitions of amphibolite-facies area, transitional zone, and granulite-facies area are based on Hiroi et al. (1991). Austhovde (698430 S, 378480 E), Innhovde (698530 S, 378120 E) and Botnnuten (708240 S, 388010 E) in Table 1 lie outside the map area, and are 53 km to the WNW, 74 km to the west, and 70 km to the SWS of Rundva˚gshetta, respectively.
Table 1. Isotopic age determinations from Prince Olav Coast and Lu¨tzow-Holm Bay, East Antarctica Lithology1
Relation2
Rb–Sr whole-rock isochron ages Gneiss Meta-trondhjemite Post-metamorphic granite Granitic rock
Syn Pre Post Post
Zone4
Sample1,5
Age (Ma)
Error (2s)
Initial ratio6
Error (2s)
Ref.7
A A A T
WR WR WR WR
683.0 1273 492.1 485
85.0 104 23.4 50
0.70564 0.70275 0.70535 0.70607
0.00075 0.00017 0.00027 0.00023
12 15 1 10
Pre Pre/Syn Syn Syn Syn Syn
Sinnan Rocks Cape Hinode Kasumi Rock Akai-Misaki (Oku-Iwa) Oku-Iwa Rock Oku-Iwa Rock Neso¨ya East Ongul Is. Breidva˚gnipa Skarvsnes
T T G G G G
WR WR WR WR WR WR
583 580 722.1 683.1 576 1131
56 23 82.8 13.2 39 81
0.70556 0.70784 0.70625 0.70471 0.71016 0.71280
0.00043 0.00059 0.00073 0.00011 0.00057 0.00094
10 7 12 12 13 12
Syn Syn Syn Syn Syn Syn Syn Syn
Cape Omega Oku-Iwa Rock Oku-Iwa Rock East Ongul Is. Skarvsnes Skallen Skallevikshalsen Rundva˚gshetta
T T T G G G G G
Bt,Pl,Ksp Bt,Pl,Ksp Bt,Hbl,FF Pl,Ksp Bt,Pl,Ksp Bt,Pl Bt,Pl Bt,FF
439.0 417.9 431 482.5 468.9 458 488 467
12.3 2.2 14 9.5 7.8 25 45 20
0.70609 0.70645 0.70718 0.71095 0.72304 0.76391 0.76155 0.74877
0.00014 0.00002 0.00038 0.00014 0.00013 0.00040 0.00090 0.00015
9 9 8 11 11 17 17 17
Pre Post Pre
Cape Hinode Kasumi Rock Oku-Iwa Rock
A T T
WR WR WR
1031 498.4 674
69 90.6 22
0.51152 0.511782 0.51192
0.00005 0.000101 0.00002
15 1 10
Pre Syn Syn
Oku-Iwa Rock Skallen Skallevikshalsen
T G G
Bt,Hbl,FF Grt,Pl,Bt Grt,Pl,Bt
36 46 62
0.511892 0.51099 0.51092
0.000040 0.00009 0.00010
8 17 17
578 474 524
167
(Continued)
POST-530 MA CHRONOLOGY OF LHC
Hbl –Bt gneiss Bt–Hbl granite Gneiss Gneiss Migmatite Gneiss Rb–Sr mineral isochron ages Granitic rock Granitic rock Pegmatitic Bt–Hbl gneiss Granitic gneiss Gneiss Grt –Bt gneiss Grt –Bt gneiss Coarse portion in Grt –Opx gneissoes block Sm –Nd whole-rock isochron ages Meta-trondhjemite Post-metamorphic granite Hbl –Bt gneiss Sm –Nd mineral isochron ages Pegmatitic Bt–Hbl gneiss Grt –Bt gneiss Grt –Bt gneiss
Locality3
168
Table 1. Continued Relation2
Locality3
Zone4
Sample1,5
Initial ratio6
Error (2s)
Coarse portion in Grt –Opx gneissose block Fine portion in Grt –Opx gneissose block CHIME U –Th –Pb ages Paragneiss Paragneiss Pelitic gneiss Pelitic gneiss SHRIMP U–Pb ages Sil–Grt –Bt gneiss Meta-trondhjemite Grt –Sil gneiss Grt –Bt gneiss Grt –Bt gneiss Grt –Bt gneiss Grt –Bt gneiss Syndeformational leucosome in metabasite Grt gneiss Hbl –Bt gneiss K –Ar ages Migmatitic pegmatite Discordant pegmatite Amphibolite Bt–Hbl gneiss Bt–Hbl gneiss Migmatitic segregation Migmatitic segregation Migmatitic pegmatite
Syn
Rundva˚gshetta
G
Grt,Bt
446
22
0.51130
0.00010
17
Syn
Rundva˚gshetta
G
Grt,Pl,Opx
511
38
0.51174
0.00009
17
Syn Syn Syn Syn
East Ongul Is. East Ongul Is. Skallen Skallen
G G G G
Mnz Mnz Mnz Mnz
533 537 560 – 500 650 – 580
10 9
3 3 6 6
Syn Pre Syn Syn Syn Syn Syn Syn
Sinnan Rocks Cape Hinode Akarui Point West– Ongul Is. Telen Telen Rundva˚gshetta Rundva˚gshetta
A A T G G G G G
Zrn Zrn Zrn Zrn Zrn Zrn Zrn Zrn
553 1017 518 532 532 1006 521 521
6 13 12 6 8 21 9 8
14 14 16 14 14 14 14 5
Syn Syn
Austhovde Innhovde
G G
Zrn Zrn
533 550
9 12
16 14
Syn Post Syn Syn Syn Syn Syn Syn
Cape Hinode Cape Hinode Cape Hinode Kasumi Rock Kasumi Rock Kasumi Rock Kasumi Rock Kasumi Rock
A A A A A A A A
Bt Ms Hbl Bt Hbl Bt Ksp Bt
480.6 499.1 526.2 492.4 492.4 481.0 463.3 472.4
Age (Ma)
Error (2s)
4.9 5.1 5.4 5.1 5.1 4.9 4.7 4.8
Ref.7
4 4 4 4 4 4 4 4
T. MIYAMOTO ET AL.
Lithology1
Syn Syn Syn Post Post Syn Syn Post Post Syn Syn Post Syn
Kasumi Rock East Ongul Is. East Ongul Is. Berrodden Berrodden Rundva˚gshetta Rundva˚gshetta Rundva˚gshetta Rundva˚gshetta Austohovde Austohovde Innhovde Botnnuten
A G G G G G G G G G G G G
Bt Bt Hbl Bt Ksp Bt Bt Bt Hbl Bt Hbl WR Bt
477.2 480 502 475.7 429.4 500.4 516.1 445.1 512.4 449.3 502.0 434.6 474.0
Syn Syn Post Post Syn Syn Syn
Cape Hinode Rundva˚gshetta Rundva˚gshetta Rundva˚gshetta Rundva˚gshetta Rundva˚gshetta Botnnuten
A G G G G G G
Bt Bt Hbl Hbl Hbl Bt Bt
480.7 514 492.4 502 509 460 471.1
4.9 15 15 5.3 4.4 5.1 5.3 5.1 5.6 4.6 5.3 21.7 4.9 0.5 (72% plateau)7 1 (total fusion)7 0.8 (96% plateau)7 1 (total fusion)8 2 (total fusion) 2 (total fusion) 0.5 (94% plateau)7
4 11 11 4 4 4 4 4 4 4 4 2 4 4 5 4 5 5 5 4
Dating results are arranged from NE (amphibolite-facies area) to SW (granulite-facies area) of Lu¨tzow-Holm Complex. Dating results published before 1980 have been compiled by Yoshida & Kaminuma (1986) and Shiraishi et al. (1989a, b). Definition of amphibolite-facies area, transitional zone and granulite-facies area is after Hiroi et al. (1991). 1 Abbrieviations: Bt, biotite; FF, felsic fraction; Grt, garnet; Hbl, hornblende; Ksp, K-feldspar; Mnz, monazite; Ms, muscovite; Opx, orthopyroxene; Pl, plagioclase; Px, pyroxene; Zrn, zircon; WR, whole rock. 2 Field relations: pre, pre-orogenic; syn, syn-orogenic; post, post-orogenic. 3 Sample localities are shown in Figure 1. 4 Subdivison of the LHC: A, amphibolite-facies area; T, transition zone; G, granulite-facies area. 5 Dated samples (minerals and whole rock). 6 References: 1, Ajishi et al. (2004); 2, Arima & Shiraishi (1993); 3, Asami et al. (1997); 4, Fraser & McDougall (1995); 5, Fraser et al. (2000); 6, Hokada & Motoyoshi (2006); 7, Kawano et al. (2005); 8, Kawano et al. (2006); 9, Nishi et al. (1999); 10, Nishi et al. (2002); 11, Shibata et al. (1985); 12, Shibata et al. (1986); 13, Shimura et al. (1998); 14, Shiraishi et al. (1994); 15, Shiraishi et al. (1995); 16, Shiraishi et al. (2003); 17, this study. 7 From Fraser McDougall (1995); plateau age defined by them are given. 8 Recalculated ages. Original data were given by Fraser & McDougall (1995).
POST-530 MA CHRONOLOGY OF LHC
Leucocratic gneiss Bt–Hbl gneiss Bt–Hbl gneiss Semi-concordant pegmatite Semi-concordant pegmatite Concordant pegmatite Bt gneiss Late-stage, linear pegmatite Mafic dyke Gneissose granite Bt–Px gneiss Ultrapotassic mafic dyke rocks Bt–Grt gneiss Ar –Ar ages Migmatitic pegmatite Bt gneiss Mafic dyke Mafic dyke Metabasite Metapelite Bt–Grt gneiss
169
170
T. MIYAMOTO ET AL.
metamorphic pressure –temperature –time (P–T –t) path for the terrane. Dating results for the LHC published prior to 1985 have been reviewed by Yoshida & Kaminuma (1986) and Shiraishi et al. (1989a, b). Almost all are Rb –Sr and K –Ar mica (biotite or muscovite) ages for gneiss, granite and pegmatite, with a few U –Pb and Pb –Pb ages of accessory minerals in pegmatites obtained by thermal ionization mass spectrometry (TIMS). With development of the sensitive high-resolution ion microprobe (SHRIMP), numerous 550 – 520 Ma U –Pb zircon ages were subsequently reported for metamorphic rocks from the LHC (Shiraishi et al. 1994, 2003; Fraser et al. 2000). Shiraishi et al. (1994) suggested that the LHC was pervasively affected by a regional early Cambrian orogenic event, caused by the suturing of East and West Gondwana between the Rayner Complex and central Dronning Maud Land (Fig. 1). Following this assertion, the LHC is considered to mostly represent metamorphosed sediments, derived predominantly from Archaean and Late Mesoproterozoic sources and deposited on the continental margins of an erstwhile Mozambique Ocean during the Neoproterozoic. This model has been supported by several recent studies (e.g. Fraser et al. 2000; Shiraishi et al. 2003; Satish-Kumar et al. 2008). In addition, the chemical Th–U – total Pb isochron method (CHIME; Suzuki & Adachi 1991) was also successfully applied to metamorphic rocks from the region, confirming the presence of the Cambrian orogeny (Asami et al. 1997; Hokada & Motoyoshi 2006). However, the latest data obtained from the southwestern LHC suggest a bimodal distribution of ages at 650 –580 Ma and 560 –500 Ma, for which a clear explanation is still lacking (Hokada & Motoyoshi 2006). Common 600 –500 Ma zircon and monazite ages from metamorphic rocks have been widely utilized in geodynamic reconstructions of the fragmented Gondwana supercontinent, especially between the LHC and adjacent terranes in South India, Sri Lanka, and Madagascar (e.g. Shiraishi et al. 1994; Braun et al. 1998; Jacobs et al. 1998; Cenki et al. 2004; Santosh et al. 2005). However, in comparison with the wealth of information JARE scientists have obtained using the above geochronological methods, there are few results available using the Rb –Sr and Sm –Nd isotopic systems. Rb –Sr and Sm– Nd mineral dating methods are useful for revealing the tectonothermal histories of metamorphic rocks, because they utilize metamorphic minerals such as garnet, amphibole and mica. By considering the conditions in which these isotopic systems are preserved in metamorphic assemblages, various stages in a single P –T–t history can be potentially resolved.
Almost all published Rb–Sr mineral dates for the LHC are model ages estimated by assuming an initial Sr ratio of 0.7115, without analysing a coexisting low Rb/Sr phase (Maegoya et al. 1968). A small number of Rb–Sr and Sm–Nd mineral isochron ages were reported by Shibata et al. (1985), Nishi et al. (1999) and Kawano et al. (2006), for metamorphic, granitic and pegmatitic rocks from the LHC (Table 1; Fig. 1). In these studies, Rb – Sr biotite – feldspar – wholerock isochron ages range from 485 to 418 Ma, with a single Sm – Nd hornblende – biotite – feldspar – whole-rock isochron age of 578 + 36 Ma. These Rb – Sr ages correspond to K – Ar mica ages from the LHC, attributed to postmetamorphic granitic and pegmatitic magmatism. The Sm – Nd age is slightly older than typical U – Th – Pb ages from zircon and monazite, and is interpreted as the age of pegmatite intrusion before regional metamorphism (Kawano et al. 2006). Comparison of dating results from the same methods will help to resolve the complex thermal history of the LHC. Assumptions of initial Sr ratios in published results also need to be tested and evaluated. In this study, we analysed rock samples from Skallen, Skallevikshalsen, and Rundva˚gshetta in the southwestern LHC, using Rb – Sr and Sm – Nd mineral dating methods. After examination of the results, we will discuss the meaning of Rb – Sr and Sm – Nd mineral ages, compare them with ages from other parts of the LHC, and discuss implications for the evolution of the Gondwana supercontinent.
Field relations and sample descriptions Three metamorphic rock samples, selected for Rb –Sr and Sm– Nd mineral geochronology, were collected from Skallen, Skallevikshalsen, and Rundva˚gshetta during JARE-39 and -40. Field relations, samples and general petrography are described first.
Garnet – biotite gneiss, Skallen Skallen is an important locality for understanding the geology of the LHC, because of the presence of a variety of highly deformed metamorphic and igneous rocks that experienced poly-stage deformation. The outcrop is divided by a prominent shear zone into northern, central and southern areas (Osanai et al. 2004). Metamorphic rocks are classified into four categories based on presumed precursors: pelitic and quartzo-feldspathic rocks, mafic and intermediate rocks, calcareous rocks, and granitic rocks (Osanai et al. 2004). Motoyoshi et al. (1989) estimated P–T conditions of metamorphism at 0.7– 0.8 GPa and 760–830 8C.
POST-530 MA CHRONOLOGY OF LHC
Garnet porphyroblasts enclose various prograde minerals such as kyanite, staurolite and sapphirine, which suggest a clockwise P–T history (Motoyoshi et al. 1989). Detailed studies on petrology, fluid inclusions, stable isotopes and Sr isotopes on a variety of rocks suggest that this region was affected by multiple stages of infiltration by fluids, which changed in composition from Cl-rich during prograde metamorphism to aqueous at cooling stage via CO2-rich condition during decompression (Satish-Kumar et al. 2006). A representative garnet –biotite gneiss (A97121905K; 05K hereafter) (Fig. 2a) was collected for mineral dating from the northern part of central Skallen. This locality is near an east –west trending fault, which divides the Skallen region into the northern and central areas. The gneiss is well-layered and associated with garnet-bearing twopyroxene–hornblende gneiss and calcareous rock layers with small lenses of garnet–sillimanite– spinel gneiss. The garnet–biotite gneiss is composed of plagioclase, alkali-feldspar and quartz, in addition to garnet and biotite (Fig. 2b). Zircon, monazite, apatite and rutile are present as accessory phases. Garnet is almandine – pyrope (Alm58 – 61Prp34 – 31Grs7 – 6Sps1 – 2) in composition. Some garnet grains are up to 3–4 mm across, whereas other minerals (quartz, plagioclase, alkalifeldspar and biotite) are generally less than 1 mm across. Numerous inclusions of fine quartz, plagioclase, alkali-feldspar, biotite and monazite were observed within larger garnet grains. Mineral assemblages and bulk-rock composition (Table 2) suggest that the protolith was pelitic, chemically modified during granulite-facies metamorphism.
Garnet – biotite gneiss, Skallevikshalsen Yoshida et al. (1976) grouped major crystalline basement rocks at Skallevikshalsen into metasediments, metabasites, marbles, skarn and associated rocks, quartzite, garnet gneissose granite, charnockites and minor intrusive rocks, similar to those exposed at Skallen. Recently, Yoshimura et al. (2004) estimated metamorphic P–T conditions at 0.60– 1.2 GPa and 770–960 8C, whereas Kawakami & Ikeda (2004) characterized relationships between peak metamorphism and subsequent ductile deformation. The sample from this outcrop is a garnet–biotite gneiss (TM990203-02A; 02A hereafter, Fig. 2c), collected from central Skallevikshalsen. The gneiss is well layered and associated in the field with charnockite, metasediments and metabasite. The garnet–biotite gneiss is composed of plagioclase, alkali feldspar, quartz and sillimanite, in addition to garnet and biotite (Fig. 2d). Zircon, monazite and apatite are present as accessory phases. Garnet
171
is almandine–pyrope (Alm56 – 60 Prp37 – 33Grs6 – 5 Sps1 – 2) in composition. Except for porphyroblastic garnets with numerous inclusions of quartz, plagioclase, alkali-feldspar and biotite, most minerals are fine grained. The intermediate and aluminous composition of 02A (Table 2), with the presence of abundant garnet and biotite, suggests a pelitic or andesitic protolith that was compositionally modified during granulite-facies metamorphism.
Garnet– orthopyroxene gneissose block, Rundva˚gshetta Basement lithologies at Rundva˚gshetta are pyroxene gneiss, garnet –biotite gneiss, garnet–sillimanite gneiss and pyroxene amphibolite (Motoyoshi et al. 1986). These are intruded by undeformed pegmatites. Yoshimura et al. (2008) observed coexisting quartz–sapphirine inclusions within garnet from sapphirine–garnet– orthopyroxene granulites, which are interlayered with garnet–biotite and garnet– sillimanite gneisses. Motoyoshi et al. (2006) confirmed extreme metamorphism with estimates of 1.0–1.1 GPa and 950–1040 8C and subsequent isothermal decompression, along a clockwise P–T path. The timing of peak UHT metamorphism is still debated, because of the presence of bimodal monazite ages (Motoyoshi et al. 2006). Fraser & McDougall (1995) calculated a cooling rate .27 8C Ma21 after peak metamorphism, and Fraser et al. (2000) estimated an exhumation rate of 3 km Ma21. A garnet–orthopyroxene gneissose block (Fig. 2e), enclosed in a layer of pyroxene amphibolite, was collected from northwestern Rundva˚gshetta (TM990201-09B; 09B hereafter). This block contains two distinct layered portions, fine grained and coarse grained, separated by a thin (c. 5 mm) quartzo-feldspathic layer. The gneissic layering is concordant with the structure of the host amphibolite. The fine-grained portion is composed of granoblastic equigranular garnet, orthopyroxene, plagioclase and quartz, up to 1 mm in grain size (Fig. 2f). Ilmenite, zircon and apatite are present in minor amounts. Garnet is almandine– pyrope (Alm61 – 65Prp26 – 23Grs11 – 10Sps2) in composition. Quartz inclusions are present in some garnet grains. The coarse-grained portion is 1 –4 cm thick. It is composed of garnet, biotite, orthopyroxene, plagioclase and quartz (Fig. 2g). Ilmenite, zircon and apatite are present as minor phases. Garnet is almandine–pyrope (Alm54 – 57 Prp37 – 33Grs7Sps2 – 3) in composition. Some garnet, orthopyroxene and biotite grains are up to 5 mm across, and have numerous inclusions of fine quartz, plagioclase and biotite. In the matrix, coarse-grained garnet and orthopyroxene are
172
T. MIYAMOTO ET AL.
Fig. 2. Polished slabs and thin sections showing representative mineral assemblages of samples analysed. (a) Hand specimen of A97121905K (05K). (b) Sample 05K, with a mineral assemblage of quartz, plagioclase, alkali-feldspar, garnet and biotite. (c) Hand specimen of TM990203-02A (02A). (d) Sample 02A with a mineral assemblage of quartz, plagioclase, alkali-feldspar, garnet, biotite and sillimanite. (e) Hand specimen of TM990201-09B. A centimetre-thick coarse-grained portion (09Bc, left) is separated from a fine-grained portion (09Bf, right) by a thin quartzo-feldspathic layer. (f) Sample 09Bf, with a mineral assemblage of quartz, plagioclase, garnet and orthopyroxene. (g) Sample 09Bc, with a mineral assemblage of quartz, plagioclase, garnet, biotite and orthopyroxene. Afs, alkali feldspar; Bt, biotite; Grt, garnet; Opx, orthopyroxene, Pl, plagioclase; Qtz, quartz; Sil, sillimanite.
POST-530 MA CHRONOLOGY OF LHC
173
Table 2. Whole-rock compositions of dating sample
wt% SiO2 TiO2 Al2O3 Fe2O3* MnO MgO CaO Na2O K2O P2O5 Total
A97121905K (05K)
TM990203-02A (02A)
TM990201-09B, fine-grained layer (09Bf)
TM990201-09B, coarse-grained layer (09Bf)
56.83 0.65 20.06 9.68 0.21 4.02 4.09 3.11 1.56 0.17 100.38
60.61 0.72 17.49 9.07 0.15 4.37 3.52 2.47 1.44 0.11 99.95
51.84 3.12 15.26 16.23 0.22 5.92 7.30 0.22 0.05 0.36 100.52
52.27 1.32 13.91 14.80 0.19 10.43 4.01 0.28 2.15 0.01 99.37
Analysed by X-ray fluorescence spectrometry (Rigaku-GF3063P). *Total iron as Fe2O3.
mantled by biotite. A protolith for the gneissose block is difficult to specify; however, the high-Mg and -Fe bulk chemistry (Table 2) suggests a mafic protolith for the fine-grained portion, whereas the coarse-grained layer was probably formed by the metasomatism associated with the injection of felsic melt veins. Fine- and coarsegrained portions were separated for dating (09Bf and 09Bc, respectively).
Analytical procedure Garnet, biotite, pyroxene and felsic fractions were separated from crushed rock samples. Plagioclase was then separated from the felsic fraction for three rock samples (05K, 02A and 09Bf). The minerals were concentrated by magnetic separation with a Frantz isodynamic separator and by gravity separation with tetrabromoethane and methylene iodide. The mineral separates were further purified by hand-picking under a binocular microscope. For Rb–Sr and Sm–Nd isotopic analysis, 100– 200 mg of each mineral fraction was used. Fractions were cleaned with acetone and then distilled water before decomposition. Conventional isotope dilution methods using mixed spikes were applied to determine Rb, Sr, Sm and Nd concentrations of the samples. A decomposed sample of each fraction was dissolved in a mixture of hydrochloric acid and oxalic acid, and passed through DOWEX 50W-X8, 200 mesh cation exchange resin to separate Rb, Sr and REE. Samarium and neodymium were dissolved by 0.2M 2-methyllactic acid, and separated from the remaining REE by passing through DOWEX 50W-X8, 200 mesh cation exchange resin adjusted
with ammonia, following the method of Kubota (1992) modified from Notsu et al. (1973). Rubidium concentrations were determined by thermal ionization mass spectrometry (TIMS) using a singlecollector HITACHI RMU5G system. Strontium and neodymium isotope compositions and total Sr, Sm and Nd concentrations were determined by TIMS using a single-collector JEOL JMS05RB system. The strontium standard of Eimer & Amend gave 87 Sr/86Sr ¼ 0.7080 + 0.0001 (1s) and standard NBS987 gave 87Sr/86Sr ¼ 0.7103 + 0.0001 (1s). The neodymium standard of JNdi-1 (GSJ (Geological Survey of Japan) standard with recommended value of 143Nd/144Nd ¼ 0.512115 + 0.000007; Tanaka et al. 2000) gave 143Nd/144Nd ¼ 0.51212 + 0.00004 (1s). Relative analytical errors in Rb, Sr, Sm and Nd concentrations are 2%, 1%, 0.5% and 1%, respectively. Analytical errors in 87Rb/86Sr and 147 Sm/144Nd ratios are both estimated to 3% and 2%, respectively. Contamination levels of Rb, Sr, Sm and Nd are under 1 10210 g, 3 10210 g, 1 10210 g and 1 10210 g per sample. The decay constants used for age calculations are 6.54 10212 a21 for 147Sm (Lugmair & Marti, 1978) and 1.42 10211 a21 for 87Rb (Steiger & Ja¨ger 1977). Rb–Sr and Sm–Nd isochrons were determined by the least-squares regression method of York (1966). The chondritic uniform reservoir (CHUR) and depleted mantle reservoir (DM) parameters for calculation of Nd model age (TCHUR and TDM) used are present 143 Nd/144Nd ¼ 0.512638 and 147Sm/144Nd ¼ 0.1966 for CHUR (Hamilton et al. 1983), and present 143Nd/144Nd ¼ 0.51315 and 147Sm/144Nd ¼ 0.2137 for DM (Peucat et al. 1988).
174
T. MIYAMOTO ET AL.
Table 3. Rb–Sr isotope composition for whole-rock and mineral fractions Rb (ppm)
Sr (ppm)
87
Rb/86Sr
87
Sr/86Sr*
A97121905K(05K), a garnet –biotite gneiss from Skallen Whole-rock (WR05K) Plagioclase (Pl05K) Biotite, 0.13–0.18 mm (Bt05K)
103 22.6 765
191 350 6.82
1.57 0.188 413
0.77384 + 0.00004 0.76513 + 0.00004 3.48534 + 0.00043
TM990203-02A(02A), a garnet –biotite gneiss from Skallevikshalsen 0.77397 + 0.00082 0.76242 + 0.00006 2.58320 + 0.00072 Coarse-grained portion in TM990201-09B(09Bc), garnet –orthopyroxene gneissose block from Rundva˚gshetta
Whole-rock (WR02A) Plagioclase (Pl02A) Biotite, 0.35–0.50 mm (Bt02A)
101 16.1 726
130 353 9.62
2.27 0.133 259
0.76842 + 0.00006 0.74911 + 0.00007 1.30513 + 0.00017 Fine-grained portion in TM990201-09B(09Bf), garnet –orthopyroxene gneissose block from Rundva˚gshetta
Whole-rock (WR09Bc) Felsic fraction (FF09Bc) Biotite, 0.25–0.35 mm (Bt09Bc) Whole-rock (WR09Bf ) Plagioclase (Pl09Bf )
129 3.3 561
1.23 0.32
129 189 20.2
71.5 175
2.9 0.188 85.2
0.0501 0.00531
0.74099 + 0.00004 0.74070 + 0.00003
*Errors given are +1s.
Results Mineral isochron and model ages Analytical results of Rb –Sr isotopic compositions for whole-rock and mineral fractions are given in Table 3. All felsic fractions from rock samples, including plagioclase, have high 87Sr/86Sr ratios (0.740–0.765). Rb –Sr isotope ratios of mineral fractions and bulk-rock sample from gneiss 05K define an isochron with an age of 458 + 25 Ma, with an initial 87Sr/86Sr ratio of 0.76391 + 0.00040 (Fig. 3). For gneiss 02A from Skallevikshalsen, the regressed line corresponds to an age of 488 + 45 Ma with an initial 87 Sr/86Sr ratio of 0.76155 + 0.00090 (Fig. 3). Rb–Sr isotopic compositions of biotite, felsic fraction and a bulk-rock sample of the coarsegrained portion of the gneissose block from Rundva˚gshetta (09Bc) approximate a regression line with a slope corresponding to an age of 467 + 20 Ma and an initial 87Sr/86Sr ratio of 0.74877 + 0.00015 (Fig. 3). Rb –Sr isotopic compositions of plagioclase and bulk-rock of the fine-grained portion (09Bf) are also listed in Table 3. Bulk-rock samples have high initial Sr isotope ratios (0.7407–0.7410) with low Rb concentrations (0.32– 1.23 ppm). Sm –Nd isotopic data for whole-rock and mineral fractions are presented in Table 4. The 147 Sm/144Nd ratios of garnet fractions fall in the range of 0.316– 1.82 for all rock samples (Table 4). Sm–Nd isotope ratios of mineral fractions from 05K include data from plagioclase, biotite and
three garnet portions that have variable 147Sm/ Nd ratios (0.316–0.583). All fractions, along with whole-rock data, define an isochron with an age of 474 + 46 Ma with an initial 143Nd/144Nd ratio ¼ 0.51099 + 0.00009 (Fig. 4). For 02A, the regressed line shows 524 + 62 Ma with an initial 143 Nd/144Nd ratio ¼ 0.51092 + 0.00010 (Fig. 4). Sm– Nd isotopic compositions of garnet, biotite and bulk-rock of 09Bc define an isochron with an age of 446 + 30 Ma and an initial 143Nd/144Nd ratio ¼ 0.51130 + 0.00012 (Fig. 4). On the other hand, Sm–Nd isotopic compositions of garnet, orthopyroxene, plagioclase and bulk-rock of 09Bf define an isochron with an age of 511 + 38 Ma with an initial 143Nd/144Nd ratio ¼ 0.51174 + 0.00009 (Fig. 4). Nd model ages were estimated at 2.82 + 0.30 Ga and 3.18 + 0.25 Ga for 05K, and 3.20 + 0.39 Ga and 3.49 + 0.32 Ga for 02A (TCHUR and TDM, respectively). Sm– Nd isotopic compositions for 09Bc and 09Bf did not define meaningful ages (e.g. TCHUR ¼ 2 4.0 Ga and TDM ¼ 12.2 Ga for 09Bf), as their Sm/Nd ratios are too high. 144
Discussion Interpretation of mineral isochron ages Previously reported mineral ages from metamorphic rocks in the LHC range from 1017 to 418 Ma (Table 1). SHRIMP U – Pb zircon and CHIME U –Th –Pb monazite ages concentrate around 550–520 Ma, whereas K –Ar and 40Ar – 39Ar ages
POST-530 MA CHRONOLOGY OF LHC
175
Fig. 3. Rb– Sr isochron diagram for metamorphic rocks (05K, 02A, 09Bc and 09Bf). Letters identifying points refer to sample code in Table 3. A mineral isochron for 05K shows an age of 458 + 25 Ma with an initial (87Sr/86Sr) ¼ 0.76391 + 0.00040 (MSWD ¼ 2.07). For 02A, a line with age of 488 + 45 Ma with an initial (87Sr/86Sr) ¼ 0.76155 + 0.00090 (MSWD ¼ 10.1) is defined. An isochron for 09Bf shows an age of 467 + 20 Ma with an initial (87Sr/86Sr) ¼ 0.74877 + 0.00015 (MSWD ¼ 0.75).
Table 4. Sm –Nd isotope composition for whole-rock and mineral fractions Sm (ppm)
Nd (ppm)
147
Sm/144Nd
143
Nd/144Nd*
A97121905K(05K), a garnet–biotite gneiss from Skallen Whole-rock (WR05K) Plagioclase (Pl05K) Biotite, 0.25–0.35 mm (Bt05K) Garnet, 0.50 –0.71 mm (Grt105K) Garnet, 0.35 –0.50 mm (Grt205K) Garnet, 0.25 –0.35 mm (Grt305K)
4.96 2.55 1.04 7.6 7.88 5.56
22.6 19.4 6.81 13.2 15.1 5.77
0.133 0.0795 0.0919 0.347 0.316 0.583
0.51145 + 0.00005 0.51123 + 0.00007 0.51126 + 0.00008 0.51205 + 0.00004 0.51194 + 0.00009 0.51284 + 0.00006
TM990203-02A(02A), a garnet –biotite gneiss from Skallevikshalsen 0.51139 + 0.00007 0.51122 + 0.00006 0.51128 + 0.00007 0.51290 + 0.00008 Coarse-grained portion in TM990201-09B(09Bc), garnet – orthopyroxene gneissose block from Rundva˚gshetta
Whole-rock (WR02A) Plagioclase (Pl02A) Biotite, 0.25–0.35 mm (Bt02A) Garnet, 0.25 –0.35 mm (Grt02A)
2.95 1.78 0.925 6.35
13 12.8 5.18 6.66
0.137 0.0844 0.108 0.577
0.51179 + 0.00009 0.51156 + 0.00007 0.51662 + 0.00003 Fine-grained portion in TM990201-09B(09Bf), garnet –orthopyroxene gneissose block from Rundva˚gshetta
Whole-rock (WR09Bc) Biotite, 0.25–0.35 mm (Bt09Bc) Garnet, 0.25 –0.35 mm (Grt09Bc)
1.63 0.17 3.85
5.36 1.32 1.28
0.184 0.0783 1.82
Whole-rock (WR09Bf ) Plagioclase (Pl09Bf ) Orthopyroxene, 0.25–0.35 mm (Opx09Bf ) Garnet, 0.25 –0.35 mm (Grt09Bf )
6.52 3.84 3.12 9.47
19.2 21.3 10.9 6.87
0.205 0.109 0.174 0.834
*Errors given are +1s.
0.51242 + 0.00004 0.51210 + 0.00006 0.51231 + 0.00008 0.51453 + 0.00006
176
T. MIYAMOTO ET AL.
Fig. 4. Sm– Nd isochron diagram for metamorphic rocks (05K, 02A, 09Bc and 09Bf). Letters identifying points refer to sample code in Table 4. A mineral isochron for 05K shows an age of 474 + 46 Ma with an initial (143Nd/ 144 Nd) ¼ 0.51099 + 0.00009 (MSWD ¼ 0.37). For 02A, the regressed line shows 524 + 62 Ma with an initial (143Nd/144Nd) ¼ 0.51092 + 0.00010 (MSWD ¼ 0.027). A mineral isochron for 09Bc shows an age of 446 + 22 Ma with an initial (143Nd/144Nd) ¼ 0.51130 + 0.00012 (MSWD ¼ 0.42); on the other hand, a mineral isochron for 09Bf shows another age of 511 + 38 Ma with an initial (143Nd/144Nd) ¼ 0.51174 + 0.00009 (MSWD ¼ 0.0014).
range between 526 and 429 Ma. The wide range in the latter ages probably records the effects of cooling after peak metamorphism, as well as the influence of intrusions of post-orogenic pegmatite and mafic dykes (Fraser & McDougall 1995). Rb–Sr and Sm–Nd mineral isochron ages obtained in this study vary over a similar range to the published K –Ar and 40Ar– 39Ar ages, even within samples. These variations can be interpreted by considering each sample. The 524 Ma Sm –Nd mineral isochron age for garnet –biotite gneiss from Skallevikshalsen (02A, Fig. 4), is comparable with published zircon and monazite ages in the LHC, whereas the Rb–Sr mineral isochron age is younger (488 Ma, Fig. 3). Garnet and biotite have high Sm/Nd and Rb/Sr ratios, respectively, and therefore control isochron ages in each isotopic systems. Radiogenic daughter elements (Nd in garnet and Sr in biotite) are less compatible than parental elements (Sm and Rb), and are therefore susceptible to re-equilibration by exchange with more daughtercompatible minerals, especially plagioclase. Exchange may be accommodated by bulk diffusion, fluid activity, and/or recrystallization of minerals (Villa 1998). If isotopic exchange is simply controlled by diffusion through a period of cooling, different ages between isotopic systems may reflect differences in closure temperatures for
isotope diffusion in the host minerals. For typical grain sizes and cooling rates in metamorphic rocks, (Tc,garnet) for the Sm –Nd system have been estimated at 700–750 8C (Hensen & Zhou 1995), much higher than biotite closure temperatures in the Rb–Sr system (350 8C, Cliff 1985). The differences in ages may reflect this, and from the mean ages for 02A an approximate cooling rate can be calculated of 11.1 –9.7 8C Ma21 over a 36 Ma interval. For garnet–biotite gneiss from Skallen (05K), Rb –Sr (458 Ma) and Sm– Nd (473 Ma) mineral isochron ages are younger than those obtained from Skallevikshalsen, despite a lack of petrographic differences between the samples (Fig. 2). However, 05K was collected close to a fault zone, and adjacent calcareous lithologies were strongly affected by CO2-rich and subsequent hydrothermal fluid infiltration during retrograde metamorphism (Satish-Kumar et al. 1998, 2006, 2008). CO2-rich fluids have potentially enhanced REE solubilities over hydrous fluids, through the formation of ionic complexes (Leroy & Turpin 1988: Ne´grel et al. 2000). It is possible that the Sm–Nd age of 05K reflects the action of CO2-rich fluid during retrograde metamorphism. If this is the case, fluids may also have affected the Rb–Sr mineral age. It is also worth noting that, for the three garnet measurements, the two coarser fractions have
POST-530 MA CHRONOLOGY OF LHC
higher Nd and lower Sm/Nd values. This could indicate contamination by inclusions of REE-rich minerals, especially monazite or apatite, which may also have affected the isochron. Age estimates from the coarse-grained (09Bc) and fine-grained (09Bf) portions of the garnet – orthopyroxene gneiss, from Rundva˚gshetta, are also variable. Both Rb –Sr (467 Ma) and Sm– Nd (446 Ma) ages for 09Bc are close to K –Ar and Ar –Ar biotite ages from other lithologies at Rundva˚gshetta (Fraser et al. 2000), including pegmatite emplaced at 450 Ma after cooling (Fraser & McDougall 1995). The coarse-grained portion of 09B is enclosed by quartzo-feldspathic selvages (Fig. 2e), a texture that could be attributed to metasomatism of the fine-grained gneiss by injections of felsic melt. Sample 09Bc has high initial Sr and low initial Nd isotopic compositions relative to the fine-grained portion, unusual for a presumed magmatic protolith. These differences are consistent with the chemically modified nature of the coarse-grained portion. The injected felsic melt was possibly related to the incongruent melting of surrounding metamorphic rocks. A 511 Ma Sm– Nd age for the fine-grained portion (09Bf) is similar to a U – Pb zircon age of 521 Ma from Rundva˚gshetta (Shiraishi et al. 1994; Fraser et al. 2000). This anhydrous, fine-grained gneiss is supposed to have experienced UHT metamorphism at 550– 520 Ma (Motoyoshi et al. 2006; Yoshimura et al. 2008). Subsequently, the portion seemed to have remained as palaeosome of the gneissose block during the 460 –450 Ma event, as discussed above. The slightly younger mean of the Sm–Nd age may represent the attainment of closure temperatures in garnet and orthopyroxene during the earlier stages of cooling. By assuming that zircon ages represent peak metamorphism and that K –Ar and 40Ar – 39Ar biotite or hornblende ages of closure temperatures, in those minerals, Fraser & McDougall (1995) derived a rapid (.27 8C Ma21) cooling path from peak metamorphism in the Rundva˚gshetta region. Assuming Tc,garnet for Sm –Nd is 750– 700 8C (Hensen & Zhou 1995), the 511 Ma age for 09Bf is consistent with this path. Fraser & McDougall (1995) obtained similar cooling rate estimates for other localities in the LHC. Our results support their conclusion that the LHC cooled quickly during uplift after peak metamorphism.
Interpretation of model ages Garnet –biotite gneiss samples 02A and 05K have high initial Sr ratios (0.761 –0.764) and low
177
initial Nd ratios (0.5109–0.5110) that reflect Archaean Nd model ages. High initial Sr ratios are also found in garnet–orthopyroxene–biotite gneiss from Skarvsnes (0.72304, Shibata et al. 1985), and the fine-grained sample 09Bf Rundva˚gshetta (0.74099), despite its extremely low Rb concentration (1.23 ppm) and low initial Nd ratio (0.51174). These isotope ratios suggest that protoliths and/or sedimentary sources derived from Archaean crust. Archaean to Proterozoic ages have been reported from detrital zircons at Rundva˚gshetta and other localities in the LHC (Shiraishi et al. 1994). The Napier Complex in Enderby Land, East Antarctica, where 2.4–2.5 Ga zircon and monazite ages are common (reviewed by Miyamoto et al. 2006), is a potential provenance region for LHC sediments. Alternatively, low Rb and high Sr isotope ratios from 09Bf may be attributed to extraction during partial melting under biotite-free conditions (Hansen et al. 2002). This sample may also have been depleted in incompatible elements during UHT granulite faciesmetamorphism, through partial melting. Meaningful model ages from 09B could not be obtained because of high Sm/Nd, as is typical of mafic rocks. The combination of high Sm/Nd and high initial Sr ratios suggests that the gneiss represents restitic material after partial melting.
Late to post-orogenic activity in the Lu¨tzow-Holm Complex and adjacent terranes The thermal history of the LHC can be broadly grouped into two stages: regional high-grade metamorphism at 550–520 Ma, and subsequent igneous and hydrothermal activity at 480–440 Ma. The former is a part of the culminating phase of Pan-African orogenesis during the assembly of East and West Gondwana, at around 530 Ma (Shiraishi et al. 1992). Many Gondwana supercontinent reconstructions place the Highland Complex of central Sri Lanka as a western extension of the LHC (e.g. Shiraishi et al. 1994). Burton & O’Nions (1990) reported Sm– Nd and Rb–Sr mineral isochron ages of 522–524 Ma and 486–501 Ma, respectively, for amphibolite and charnockite from central Sri Lanka. These ages correspond well to mineral isochron ages obtained in this study. As such, these ages have regional significance, providing potential guideposts for a widespread cooling history following the assembly of the Gondwana supercontinent. The younger group of ages also has tectonic significance. In the northeastern LHC, 439– 418 Ma ages reported for granites represent the final episode of igneous activity following regional
178
T. MIYAMOTO ET AL.
Fig. 5. CaO–Al2O3 discrimination diagram for ultrapotassic igneous rocks (K2O . 3 wt% and MgO . 3 wt%) intruded about 480–450 Ma in East Antarctica (modified after Miyamoto et al. 2000). Data compiled from Sheraton & England (1980), Kojima et al. (1982); Sandiford & Wilson (1983); Shiraishi et al. (1983), Sheraton (1985), Sakiyama et al. (1988), Mikhalsky et al. (1992), Arima & Shiraishi (1993), Oba & Shiraishi (1994), Ikeda et al. (1995), Hoch & Tobschall (1998) and Miyamoto et al. (2000). Discrimination curves and groupings are after Foley et al. (1987). Ultrapotassic rocks from the Napier Complex in Enderby Land and the Rayner Complex in the southern Prince Charles Mountains of Mac Robertson Land (filled symbols) plot in the Group I field. Ultrapotassic rocks from Mac Robertson Land contain more TiO2 than those from Enderby Land. Potassic and ultrapotassic rocks from Dronning Maud Land (open symbols) plot mainly in Group IV.
metamorphism, although Kawano et al. (2006) have recently suggested that they may represent cooling ages rather than crystallization ages. Coeval pegmatitic and ultrapotassic intrusions are also found in the southwestern LHC (Saito & Sato 1964; Shibata et al. 1985; Arima & Shiraishi 1993; Fraser & McDougall 1995). Ultrapotassic igneous rocks emplaced during 480– 450 Ma occur in adjacent terranes, including the Yamato –Belgica and Sør-Rondane Complexes in Dronning Maud Land, the Napier Complex in Enderby Land, and the southern Prince Charles Mountains (Rayner Complex) in MacRobertson Land (e.g. Sheraton & England 1980; see also Fig. 5). Arima & Shiraishi (1993) and Ikeda et al. (1995) have suggested that the ultrapotassic magmatism in eastern Dronning Maud Land represents the waning stage of orogenesis produced during the amalgamation of East and West Gondwana. Geochemical data compiled by Miyamoto et al. (2000) demonstrate differences between ultrapotassic igneous rocks from Enderby Land –MacRobertson Land and those from Dronning Maud Land (Fig. 5). Samples from Enderby Land and MacRobertson Land have characteristics of magmas generated in association
with stabilized continental crust, whereas those from Dronning Maud Land preserve signatures of subducted crustal material, consistent with magma generation subsequent to ocean closure and intercontinental collision. Post-480 Ma magmatic and thermal activity is spread over a wide area of East Antarctica, and its presence in both stabilized continental blocks and late Neoproterozoic suture zones is part of the post-orogenic stabilization of Gondwana.
Conclusion Metamorphic rocks from the southern LHC have Rb –Sr and Sm–Nd mineral ages that fall into two ranges, 524– 488 Ma and 474–446 Ma, which represent different stages of the end of a regional orogenic event. An age of 524 Ma is indistinguishable from U –Th– Pb ages of zircon and monazite, assumed to have grown at or near the peak of metamorphism. Younger Sm –Nd (511 Ma) and Rb –Sr (488 Ma) mineral ages record stages of cooling after peak metamorphism. Further young mineral ages of 474–446 Ma correspond to the
POST-530 MA CHRONOLOGY OF LHC
timing of hydrothermal or subsequent granitic and pegmatitic magmatism. The ages represent the waning stages of continental development after the late Neoproterozoic amalgamation of East and West Gondwana. We express our sincere thanks to the members of JARE-40 and crew of the icebreaker Shirase, especially Y. Ohashi, K. Maki, S. Harigai, T. Takei, H. Yamauchi and K. Shiraishi, who supported us in the helicopter operation. We would like to thank E. S. Grew, H. Ishizuka, M. Owada, Y. Hiroi, T. Kawasaki, T. Tsunogae, T. Ikeda, T. Hokada, A. Kamei, N. Nakano and K. Sato for their constructive comments and discussions during preparation of this manuscript. Constructive critical reviews from Y. Hiroi, H. Kagami, T. Nakajima and M. Flowerdew are greatly appreciated. T.M.’s research was supported financially by a Grant-in-Aid for Scientific Research (c) (No. 19540484) from the Ministry of Education, Science, Sports and Culture, Japan. M.S.-K. acknowledges JSPS Grant-in-Aid No. 18740319.
References A JISHI , H., K AWANO , Y., K AWAKAMI , T. & I KEDA , T. 2004. Geochronological study of post metamorphic granite from Kasumi Rock, Lu¨tzow-Holm Complex, East Antarctica. Polar Geoscience, 17, 35–44. A RIMA , M. & S HIRAISHI , K. 1993. Geochemical characteristics of metamorphosed high K/Na dykes in eastern Queen Maud Land, Antarctica: ultrapotassic igneous activity linked to Pan African orogeny. Proceedings of NIPR Symposium on Antarctic Geosciences, 6, 103–115. A SAMI , M., S UZUKI , K. & A DACHI , M. 1997. Th, U and Pb analytical data and CHIME dating of monazites from metamorphic rocks of the Rayner, Lu¨tzowHolm, Yamato–Belgica and Sør Rondane Complexes, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 10, 130– 152. B RAUN , I., M ONTEL , J.-M. & N ICOLLET , C. 1998. Electron microprobe dating of monazites from highgrade gneisses and pegmatites of the Kerala Khondalite Belt, southern India. Chemical Geology, 146, 65– 85. B URTON , K. W. & O’N IONS , R. K. 1990. The timescale and mechanism of granulite formation at Kurunegala, Sri Lanka. Contributions to Mineralogy and Petrology, 106, 66–89. C ENKI , B., B RAUN , I. & B ROCKER , M. , 2004. Evolution of the continental crust in the Pan-African mobile belt of southernmost India: evidence from Nd isotope mapping combined with U– Pb and Rb–Sr geochronology. Precambrian Research, 134, 275– 292. C LIFF , R. A. 1985. Isotopic dating in metamorphic belts. Journal of the Geological Society, London, 142, 97– 110. F OLEY , S. F., V ENTTURELLI , G., G REEN , D. H. & T OSCANI , L. 1987. The ultrapotassic rocks: characteristics, classification and constraints for petrogenetic models. Earth-Science Reviews, 24, 81– 134.
179
F RASER , G. & M C D OUGALL , U. 1995. K/Ar and 40 Ar/39Ar mineral ages across the Lu¨tzow-Holm Complex, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 8, 137– 159. F RASER , G., M C D OUGALL , U., E LLIS , D. J. & W ILLIAMS , I. S. 2000. Timing and rate of isothermal decompression in Pan-African granulites from Rundva˚gshetta, East Antarctica. Journal of Metamorphic Geology, 18, 441–454. H AMILTON , P. J., O’N IONS , R. K., B RIDGWATER , D. & N UTMAN , A. 1983. Sm– Nd studies of Archean metasediments and metavolcanics from West Greenland and their implications for the Earth’s early history. Earth and Planetary Science Letters, 62, 263– 272. H ANSEN , E., A HMED , K. & H ARLOV , D. E. 2002. Rb depletion in biotites and whole rocks across an amphibolite to granulite facies transition zone, Tamil Nadu, South India. Lithos, 64, 29– 47. H ENSEN , B. J. & Z HOU , B. 1995. Retention of isotopic memory in garnets partially broken down during an overprinting granulite-facies metamorphism: implications for the Sm–Nd closure temperature. Geology, 23, 225– 228. H IROI , Y., S HIRAISHI , K. & M OTOYOSHI , Y. 1991. Late Proterozoic paired metamorphic complexes in East Antarctica, with special reference to the tectonic significance of ultramafic rocks. In: T HOMSON , M. R. A., C RAME , J. A. & T HOMSON , J. W. (eds) Geological Evolution of Antarctica. Cambridge University Press, Cambridge, 83– 87. H OCH , M. & T OBSCHALL , H. J. 1998. Minettes from Schirmacher Oasis, East Antarctica—indicators of an enriched mantle source. Antarctic Sciences, 10, 476– 486. H OKADA , T. & M OTOYOSHI , Y. 2006. Electron microprobe technique for U–Th– Pb and REE chemistry of monazite, and its implications for pre-, peak- and post-metamorphic events of the Lu¨tzow-Holm Complex, East Antarctica. Polar Geoscience, 19, 118– 151. I KEDA , Y., S HIRAISHI , K. & K AGAMI , H. 1995. Geochemical characteristics of mafic dike rocks from the Sør Rondane Mountains, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 8, 49–64. J ACOBS , J., F ANNING , C. M., H ENJES -K UNST , F., O LESCH , M. & P AECH , H.-J. 1998. Continuation of the Mozambique Belt into East Antarctica: Grenvilleage metamorphism and polyphase Pan-African highgrade events in Central Dronning Maud Land. Journal of Geology, 106, 385 –406. K AWAKAMI , T. & I KEDA , T. 2004. Timing of ductile deformation and peak metamorphism in Skallevikshalsen, Lu¨tzow-Holm Complex, East Antarctica. Polar Geoscience, 17, 1 –11. K AWANO , Y., M ENO , A., N ISHI , N. & K AGAMI , H. 2005. Geochemistry of the pre/syn-metamorphic granite in the Ongul Islands, East Antarctica. Polar Geoscience, 18, 114– 129. K AWANO , Y., N ISHI , N. & K AGAMI , H. 2006. Rb– Sr and Sm–Nd mineral isochron ages of a pegmatitic gneiss from Oku-iwa Rock, Lu¨tzow-Holm Complex, East Antarctica. Polar Geoscience, 19, 109– 117.
180
T. MIYAMOTO ET AL.
K OJIMA , H., Y ANAI , K. & N ISHIDA , T. 1982. Geology of the Belgica Mountains. Memoirs of National Institute of Polar Research, Special Issue, 21, 32– 46. K UBOTA , M. 1992. Sm–Nd model ages of aeolian dusts in Japan and Hawaii and volcanic rocks on the greenstone belt in the Tanzanian Shield. MS thesis, Kyushu University, Fukuoka [in Japanese with English abstract]. L EROY , J. L. & T URPIN , L. 1988. REE, Th and U behaviour during hydrothermal and supergene processes in a granitic environment. Chemical Geology, 68, 239– 251. L UGMAIR , G. W. & M ARTI , K. 1978. Lunar initial 143 Nd/144Nd: differential evolution of the lunar crust and mantle. Earth and Planetary Science Letters, 39, 349– 357. M AEGOYA , T., N OHDA , S. & H AYASE , I. 1968. Rb– Sr dating of the gneissic rocks from the East Coast of Lu¨tzow-Holm Bay, Antarctica. Memoirs of Faculty of Science, Kyoto University, Series of Geology and Mineralogy, 35, 131–138. M IKHALSKY , E. V., A NDRONIKOV , A. V. & B ELIATSKY , B. V. 1992. Mafic igneous suites in the Lambert Rift Zone. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Science. Terra, Tokyo, 173–178. M IYAMOTO , T., G REW , E. S., S HERATON , J. W. ET AL . 2000. Lamproite dykes in the Napier Complex at Tonagh Island, Enderby Land, East Antarctica. Polar Geoscience, 13, 41– 59. M IYAMOTO , T., Y OSHIMURA , Y., D UNKLEY , D. J., M OTOYOSHI , Y. & C ARSON , C. J. 2006. Mineral ages for multi isotope system in phlogopite-bearing pyroxene granulite and felsic gneiss, the Howard Hills, Enderby Land, East Antarctica: Possible Proterozoic tectonothermal events in the Napier Complex. Polar Geoscience, 19, 152–172. M OTOYOSHI , Y., M ATSUEDA , H., M ATSUBARA , S., S ASAKI , K. & M ORIWAKI , K. 1986. Geological map of Rundva˚gskollane and Rundva˚gshetta, Antarctica. Antarctic Geological Map Series, Sheet 24. National Institute of Polar Research, Tokyo. M OTOYOSHI , Y., M ATSUBARA , S. & M ATSUEDA , H. 1989. P –T evolution of the granulite-facies rocks of the Lu¨tzow-Holm Bay region, East Antarctica. In: D ALY , J. S., C LIFF , R. A. & Y ARDLEY , B. W. D. (eds) Evolution of Metamorphic Belts. Geological Society, London, Special Publications, 43, 325 –329. M OTOYOSHI , Y., H IROI , Y., S ATISH -K UMAR , M. & H OKADA , T. 2006. Timing of UHT metamorphic event at Rundva˚gshetta, Lu¨tzow-Holm Complex, East Antarctica. Abstracts of the 113th Annual Meeting of the Geological Society of Japan, 139. N E´ GREL , Ph., G UERROT , C., C OCHERIE , A., A ZAROUAL , M., B RACH , M. & F OUILLAC , Ch. 2000. Rare earth elements, neodymium and strontium isotopic systematics in mineral water: evidence from the Massif Central, France. Applied Geochemistry, 15, 1345–1367. N ISHI , N., K AWANO , Y. & K AGAMI , H. 1999. Preliminary result for the Rb– Sr mineral isochron ages of granitic rocks from Cape Omega and Oku-iwa Rock, Prince Olav Coast, East Antarctica. Polar Geoscience, 12, 157 –165.
N ISHI , N., K AWANO , Y. & K AGAMI , H. 2002. Rb–Sr amd Sm–Nd isotopic geochronology of the granitoid and hornblende biotite gneiss from Oku-iwa Rock in the Lu¨tzow-Holm Complex, East Antarctica. Polar Geoscience, 15, 46–65. N OTSU , K., M ABUCHI , H., Y OSHIOKA , O., M ATSUDA , J. & O ZIMA , M. 1973. Evidence of the extinct nuclide 146 Sm in ‘Juvinas’ achondrite. Earth and Planetary Science Letters, 19, 29–36. O BA , T. & S HIRAISHI , K. 1994. The emplacement pressure of syenite estimated from the stability field of amphibole from the Yamato Mountains, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 7, 60–68. O SANAI , Y., T OYOSHIMA , T., O WADA , M., ET AL . 2004. Geological map of Skallen (revised edition), Antarctica. Antarctic Geological Map Series, Sheet 39. National Institute of Polar Research, Tokyo. P EUCAT , J. J., V IDAL , P., B ERNARD -G RIFFITHS , J. & C ONDIE , K. C. 1988. Sr, Nd and Pb isotopic systematics in the Archean low- to high-grade transition zone of southern India: syn-accretion vs. post-accretion granulites. Journal of Geology, 97, 537– 550. S AITO , N. & S ATO , K. 1964. On the age of euxenite from Antarctica. In: A DIE , R. J. (ed.) Antarctic Geology. North Holland, Amsterdam, 590 –596. S AKIYAMA , T., T AKAHASHI , Y. & O SANAI , Y. 1988. Geological and petrological characters of the plutonic rocks in the Lunckeryggen– Brattnipene region, Sør Rondane Mountains, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 2, 80–95. S ANDIFORD , M. & W ILSON , C. J. L. 1983. The geology of the Fyfe Hills–Khmara Bay region, Enderby Land. In: O LIVER , R. L., J AMES , P. R. & J AGO , J. B. (eds) Antarctic Earth Science. Australian Academy of Science, Canberra, A.C.T., 16–19. S ANTOSH , M., C OLLINS , A. S., M ORIMOTO , T. & Y OKOYAMA , K. 2005. Depositional constraints and age of metamorphism in southern India: U – Pb chemical (EMPA) and isotopic (SIMS) ages from the Trivandrum Block. Geological Magazine, 142, 255–268. S ATISH -K UMAR , M., Y OSHIDA , M., W ADA , H., N IITSUMA , N. & S ANTOSH , M. 1998. Fluid flow along microfractures in calcite from a marble from East Antarctica: evidence from gigantic (21‰) oxygen isotopic zonation. Geology, 26, 251– 254. S ATISH -K UMAR , M., H ERMANN , J., T SUNOGAE , T. & O SANAI , Y. 2006. Carbonation of Cl-rich scapolite boudins in Skallen, East Antarctica: evidence for changing fluid condition in the continental crust. Journal of Metamorphic Geology, 24, 241 –261. S ATISH -K UMAR , M., M IYAMOTO , T., H ERMANN , J., K AGAMI , H., O SANAI , Y. & M OTOYOSHI , Y. 2008. Pre-metamorphic carbon, oxygen and strontium isotope signature of high-grade marbles from the Lu¨tzow-Holm Complex, East Antarctica: apparent age constraints of carbonate deposition. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctic: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 147–164.
POST-530 MA CHRONOLOGY OF LHC S HERATON , J. W. 1985. Chemical analyses of rocks from East Antarctica: Part 2. Bureau of Mineral Resources, Geology and Geophysics, Australia, Record 1985/12. S HERATON , J. W. & E NGLAND , R. N. 1980. Highly potassic mafic dykes from Antarctica. Journal of the Geological Society of Australia, 27, 129–135. S HIBATA , K., Y ANAI , K. & S HIRAISHI , K. 1985. Rb–Sr mineral isochron ages of metamorphic rocks around Syowa station and from the Yamato Mountains, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 37, 164 –171. S HIBATA , K., Y ANAI , K. & S HIRAISHI , K. 1986. Rb–Sr whole-rock ages of metamorphic rocks from eastern Queen Maud Land, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 43, 133–148. S HIMURA , T., F RASER , G. L., T SUCHIYA , N. & K AGAMI , H. 1998. Genesis of the migmatites of Breidva˚gnipa, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 53, 109–136. S HIRAISHI , K., A SAMI , M. & K ANAYA , H. 1983. Petrochemical character of the syenitic rocks from the Yamato Mountains, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 28, 183–197. S HIRAISHI , K., H IROI , Y. & M OTOYOSHI , Y. 1989a. Geological map of Lu¨tzow-Holm Bay, Antarctica. Antarctic Geological Map Series, Sheet 12. National Institute of Polar Research, Tokyo. S HIRAISHI , K., H IROI , Y. & M OTOYOSHI , Y. 1989b. Geological map of Prince Olav Coast, Antarctica. Antarctic Geological Map Series, Sheet 13. National Institute of Polar Research, Tokyo. S HIRAISHI , K., H IROI , Y., E LLIS , D. J., F ANNING , C. M., M OTOYOSHI , Y. & N AKAI , Y. 1992. The first report of a Cambrian orogenic belt in East Antarctica—an ion microprobe study of the Lutzow-Holm Complex. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Sciences. Terra, Tokyo, 67– 73. S HIRAISHI , K., E LLIS , D. J., H IROI , Y., F ANNING , C. N., M OTOYOSHI , Y. & N AKAI , Y. 1994. Cambrian orogenic belt in East Antarctica and Sri Lanka: implications for Gondwana assembly. Journal of Geology, 102, 47–65. S HIRAISHI , K., K AGAMI , H. & Y ANAI , K. 1995. Sm–Nd and Rb –Sr isochron ages for meta-trondhjemites from
181
Cape Hinode, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 8, 130– 136. S HIRAISHI , K., H OKADA , T., F ANNING , C. N., M ISAWA , K. & M OTOYOSHI , Y. 2003. Timing of thermal events in eastern Dronning Maud Land, East Antarctica. Polar Geoscience, 16, 76– 99. S TEIGER , R. H. & J A¨ GER , E. 1977. Subcommission on geochronology: convention on the use of decay constants in geo- and cosmo-chronology. Earth and Planetary Science Letters, 36, 359–362. S UZUKI , K. & A DACHI , M. 1991. Precambrian provenance and Silurian metamorphism of the Tsubonosawa paragneiss in the South Kitakami terrane, Northeast Japan, revealed by Th–U– total Pb chemical isochron ages of monazite, zircon and xenotime. Geochemical Journal, 25, 357–376. T ANAKA , T., T OGASHI , S., K AMIOKA , H. ET AL . 2000. JNdi-1: a neodymium isotopic reference in consistency with La Jolla neodymium. Chemical Geology, 168, 279– 281. V ILLA , I. M. 1998. Isotopic closures. Terra Nova, 10, 42–47. Y ORK , D. 1966. Least-squares fitting of a straight line. Canadian Journal of Physics, 44, 1079– 1086. Y OSHIDA , M., Y OSHIDA , Y., A NDO , H., I SHIKAWA , T. & T ATSUMI , T. 1976. Geological map of Skallen, Antarctica. Antarctic Geological Map Series, Sheet 9. National Institute of Polar Research, Tokyo. Y OSHIDA , Y. & K AMINUMA , K. 1986. Science in Antarctica, 5, Earth Sciences. National Institute of Polar Research, Tokyo [in Japanese]. Y OSHIMURA , Y., M OTOYOSHI , Y., M IYAMOTO , T., G REW , E. S., C ARSON , C. J. & D UNKLEY , D. J. 2004. High-grade metamorphic rocks from Skallevikshalsen in the Lu¨tzow-Holm Complex, East Antarctica: metamorphic conditions and possibility of partial melting. Polar Geoscience, 17, 57–87. Y OSHIMURA , Y., M OTOYOSHI , Y. & M IYAMOTO , T. 2008. Sapphirine þ quartz association in garnet: implication for ultrahigh-temperature metamorphism in Rundva˚gshetta, Lu¨tzow-Holm Complex, East Antarctica. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctic: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 377–390.
Elastic properties of high-grade metamorphosed igneous rocks from Enderby Land and eastern Dronning Maud Land, Antarctica: evidence for biotite-bearing mafic lower crust M. ISHIKAWA, E. SHINGAI & M. ARIMA Graduate School of Environment and Information Sciences, Yokohama National University, Tokiwadai 79-7, Hodogaya-ku, Yokohama 2408501, Japan (e-mail:
[email protected]) Abstract: Ultrasonic measurements of P-wave velocity (Vp) and S-wave velocity (Vs) were conducted at high pressures up to 1.0 GPa and high temperatures up to 400 8C for ultrahightemperature (UHT) metamorphosed rocks from Mount Riiser-Larsen, in the Archaean Napier Complex of Enderby Land. The results at 1.0 GPa and 400 8C are Vp ¼ 7.17 km s21, Vs ¼ 4.24 km s21, Vp/Vs ¼ 1.69, Poisson’s ratio (s) ¼ 0.23 for pyroxenite (SiO2 ¼ 44.2 wt%, density (r) ¼ 3.41 g cm23); Vp ¼ 6.93 km s21, Vs ¼ 3.81 km s21, Vp/Vs ¼ 1.82, s ¼ 0.28 for mafic granulite (SiO2 ¼ 52.2 wt%, r ¼ 3.02 g cm23); Vp ¼ 6.88 km s21, Vs ¼ 3.72 km s21, Vp/Vs ¼ 1.85, s ¼ 0.29 for mafic granulite (SiO2 ¼ 49.5 wt%, r ¼ 2.88 g cm23); and Vp ¼ 6.17 km s21, Vs ¼ 3.59 km s21, Vp/Vs ¼ 1.72, s ¼ 0.24 for orthopyroxene-bearing felsic gneiss (SiO2 ¼ 65.4 wt%, r ¼ 2.68 g cm23). Vp and Vp/Vs of these UHT rocks are not comparable with the previously proposed seismic velocity model (Vp ¼ 6.56 km s21, Vp/Vs ¼ 1.70) for the lower crust beneath the Mizuho Plateau of eastern Dronning Maud Land. Combining the available measured velocity and density data with the seismic velocity profile defined for the Mizuho Plateau, we suggest that relatively low Vp and Vs characteristics of the lower crust beneath the Mizuho Plateau may be attributed to higher abundance of biotite in the mafic lower crustal rocks. It is proposed that the biotite-bearing lower crustal rocks were formed by metasomatic processes associated with Pan-African orogeny.
In recent years, great progress has been made in the understanding of the Earth’s continental crust by means of geological mapping and geophysical surveys of large areas. In addition, rock and mineral physics has become a critical tool to provide data essential for understanding the structure, composition, rheological behaviour, and physical state of the deeper crust. Knowledge of the high-pressure and high-temperature physical properties of deepseated crustal rocks is critical for studies of crustal structure and composition as well as for understanding the deep crustal geodynamics of continent and the evolution of orogenic belts (e.g. Christensen & Mooney 1995; Mueller 1995; Mueller & Massonne 2001). Today seismic experiments show the velocity structure and lateral changes within the crust. These variations are inferred to reflect rock type, temperature difference, partial melting, structure of magma chambers and fluid infiltration within the crust. However, without an understanding of the elastic properties of the crustal rocks, the interpretation of the seismic structure in terms of lithology and composition is ambiguous. As a part of the geoscientific work performed by the Japanese Antarctic Research Expedition (JARE), Dronning Maud Land and Enderby Land were chosen as ideal fields for the ‘Structure and Evolution of the East Antarctic Lithosphere
Project (SEAL)’ (Fig. 1). The SEAL project is a comprehensive and integrated study of the continental crust comprising geological mapping, petrological, geochemical and geochronological studies, and geophysical surveys. Seismic explosion experiments were conducted along the traverse route from Syowa Station to Mizuho Station by JARE-41 in 1999– 2000 (Miyamachi et al. 2001; Tsutsui et al. 2001a, b); this is the same seismic traverse route as for JARE-21, carried out during 1980– 1981 (Ikami et al. 1984; Ito & Ikami 1984, 1986; Ito & Kanao 1995). The traverse route is located in the southeastern part of the Cambrian Lu¨tzow-Holm Complex (c. 550–520 Ma; Shiraishi et al. 1997), and 300 km west of the Archaean Napier Complex (Sheraton et al. 1987). Determination of the elastic properties of highgrade metamorphic rocks from Dronning Maud Land and Enderby Land is essential for the interpretation of seismic velocity models (e.g. Yoshii et al. 2004) derived from the SEAL project. Yukutake & Ito (1984) measured P-wave velocities of metamorphic rock from Ongul Island at pressures up to 0.44 GPa. However, the confining pressure of 0.44 GPa is substantially smaller than that of lower crustal depth. Previously, we have developed techniques to perform ultrasonic P-wave velocity measurements at pressures and
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 183 –194. DOI: 10.1144/SP308.9 0305-8719/08/$15.00 # The Geological Society of London 2008.
184
M. ISHIKAWA ET AL.
Fig. 1. Simplified geological map of eastern Dronning Maud Land and Enderby Land in East Antarctica showing ages (Ma) of the Lu¨tzow-Holm Complex, Rayner Complex and Napier Complex (after Shiraishi et al. 1994, 1997; Hokada et al. 2003). The traverse route of seismic explosion experiment during the SEAL project at the Mizuho Plateau is indicated. The 550– 520 Ma Cambrian orogeny is recognized in the Lu¨tzow-Holm Complex and the Rayner Complex, and the Cambrian orogenic belt is bounded by the Archaean Napier Complex on the east and the late Proterozoic Yamato–Belgica Complex in the west.
temperatures representative of continental lower crust (Shingai et al. 2001; Kitamura et al. 2001, 2003; Nishimoto et al. 2005), and these techniques have been extended to a higher temperature of 800 8C (Kono et al. 2004, 2006). In earlier work we measured only P-wave velocities of the UHT meta-igneous rocks exposed in the Archaean Napier Complex of Enderby Land (Shingai et al. 2001). The primary purpose of this study is to present data for both P- and S-wave velocities (Vp and Vs) and elastic parameters (Vp/Vs, Poisson’s ratio and acoustic impedance) at high pressures up to 1.0 GPa and high temperatures up to 400 8C. The Vp vs. density plots and Vp vs. SiO2 content plots are also shown with results of Kitamura et al. (2001) for meta-igneous rocks from the Lu¨tzow-Holm Complex. Finally, the highpressure and high-temperature rock velocities corresponding to lower crustal depths are combined with the seismic velocity model proposed by JARE-41 (Yoshii et al. 2004) to discuss the lithology, chemical composition and evolution of the lower crust beneath the Mizuho Plateau.
Rock specimens and experimental technique The Archaean Napier Complex of Enderby Land, 300 km west of Lu¨tzow-Holm Bay, is represented by ultrahigh-temperature (UHT) metamorphism (Harley & Hensen 1990). Based on the detailed geological mapping during JARE-38, Ishikawa et al. (2000) and Ishizuka et al. (1998) showed that
UHT meta-igneous rocks of Mount Riiser-Larsen are bimodal in SiO2 composition. For the ultrasonic P- and S-wave velocity measurements, we selected four rock samples having homogeneous texture from Mount Riiser-Larsen as the rocks representing major lithologies. These rocks have whole-rock compositions ranging from felsic (66 wt% SiO2) through mafic (52–49 wt% SiO2) to ultramafic (44 wt% SiO2). The rock samples subjected to ultrasonic measurements are pyroxenite (sample MI97012402), mafic granulites (MI97012710, MI96122602) and orthopyroxene-bearing felsic gneiss (MI97012401). Detailed sample locations were given in the geological map of Shingai et al. (2001). The whole-rock compositions were analysed by the X-ray fluorescence (XRF) method at the National Institute of Polar Research (NIPR), Tokyo, Japan. The whole-rock compositions and mineral assemblages are listed in Table 1. Pyroxenite (MI97012402) is a homogeneous rock with a granoblastic texture. It contains orthopyroxene (51.3 vol.%) and clinopyroxene (31.3 vol.%) with small amounts of plagioclase (4.0 vol.%) and biotite (2.6 vol.%). The rock is classified as websterite. The XRF analysis of pyroxenite shows basic composition with a SiO2 content of 44.18 wt.%. Grain size of mineral constituents ranges from 0.8 to 1.0 mm. Orthopyroxene-bearing felsic gneiss is a homogeneous rock with medium grain size (c. 1.0 mm). No foliation or lineation was found. Pyroxenes and plagioclase are the predominant mineral constituents. This rock exhibits granoblastic polygonal texture. The XRF analysis of the gneiss shows a SiO2 content of 65.42 wt%
ELASTIC PROPERTIES OF META-IGNEOUS ROCKS
185
Table 1. Chemical and physical properties for the rock samples Sample no.: Rock type: Bulk chemistry (wt.%) SiO2 TiO2 Al2O3 Fe2O3 MnO MgO CaO Na2O K2O P2O5 Total Mode (vol.%) Opx Cpx Pl Qtz Bt Opq Density (g cm23)
M97012710 Pyroxene granulite
MI97012401 Opx felsic gneiss
MI96122602 Pyroxene granulite
MI97012402 Pyroxenite
52.23 1.17 13.52 14.17 0.15 6.28 9.22 1.52 0.34 0.06 98.67
65.42 0.44 15.96 3.87 0.06 1.53 4.05 4.13 1.90 0.09 97.46
49.50 0.72 14.20 14.00 0.18 8.45 9.14 2.16 0.33 0.05 98.74
44.18 1.69 5.25 20.77 0.22 15.37 9.99 0.00 1.03 0.06 98.48
24.6 19.5 52.8 – 0.4 2.7 3.02
6.5 8.0 81.5 2.5 – 1.5 2.68
17.0 14.0 61.5 – 1.0 6.5 2.88
51.3 31.3 4.0 – 2.6 5.8 3.41
and tonalitic in composition. Mafic granulites (MI97012710 and MI96122602) are homogeneous medium- to coarse grained rocks, with grain size of 1.5–2.5 mm and a granoblastic texture. The main mineral constituents are plagioclase (52.8 and 61.5 vol.%), orthopyroxene (24.6 and 17.0 vol.%) and clinopyroxene (19.5 and 14.0 vol.%). The experimental procedures have been reported by Shingai et al. (2001), Kitamura et al. (2003), and Nishimoto et al. (2005) and will be only briefly described here. Measurements of ultrasonic wave velocities in the four rock samples were carried out up to 1.0 GPa and 400 8C with a piston-cylinder-type apparatus of 34 mm bore at Yokohama National University. Talc and pyrophyllite were used as the pressure-transmitting medium. Pressures were calibrated by Vp measurements of the high –low quartz transition (Kono et al. 2008). High temperatures were achieved using a graphite heater and were monitored by a Pt–Rh13 thermocouple, which was placed upon the top end of the core rock sample. Rock samples were cut into a 14 mm diameter core with a length of about 12 mm. Both front faces of the core sample were polished. The length was measured with a micrometer and the uncertainty is estimated to be about +0.05 mm. These core samples were then oven-dried for 24 h. Pure-mode P- or S-wave lithium niobate transducers were placed on both front faces of a core sample. Travel times along a core sample of known length were measured using the pulse transmission technique. A more
detailed description of the determination of P-wave travel times has been given by Shingai et al. (2001). Travel times for S waves were measured using the phase comparison method employing a sinusoidal pulse (4 MHz for shear waves). Electric pulses were applied to the lithium niobate transducer to produce elastic compressional- or shear-wave pulses. Another lithium niobate transducer received the resulting elastic waves reconverted into electrical signals. The digitized data are saved on a hard drive for the evaluation of the travel times. Electrical wave forms were measured 4096 times and averaged for each pressure –temperature condition. The Vp and Vs values reported here represent these average values. Errors of the present ultrasonic velocity measurements are estimated to be less than 0.1 km s21.
Experimental results P- and S-wave velocities and elastic parameters (Vp/Vs and Poisson’s ratio) were determined at high pressures and high temperatures up to 1.0 GPa and 400 8C. We first measured P- or S-wave velocity at room temperature and various pressures with a step rate of 0.1 GPa during pressurization from 0.1 to 1.0 GPa. During decompression from 1.0 to 0.1 GPa, P- or S-wave velocity at various temperatures (from room temperature to higher temperatures up to 400 8C) was measured at constant pressure with a 0.1 GPa pressure interval. P- and S-wave velocities
186
M. ISHIKAWA ET AL.
Table 2. Compressional- and shear-wave velocities (Vp, Vs), velocity ratios (Vp/Vs), Poisson’s ratio (s) at 1.0 GPa and 400 8C and 0.6 GPa and 25 8C Lithology 1.0 GPa, 400 8C Mafic granulite (52.2%) Mafic granulite (49.5%) Opx felsic gneiss Pyroxenite 0.6 GPa, 25 8C Mafic granulite (52.2%) Mafic granulite (49.5%) Opx felsic gneiss Pyroxenite
Vp (km s21)
Vs (km s21)
Vp/Vs
s
6.93 6.88 6.17 7.17
3.81 3.72 3.59 4.24
1.82 1.85 1.72 1.69
0.28 0.29 0.24 0.23
7.18 7.23 6.52 7.64
3.87 3.82 3.64 4.44
1.86 1.89 1.79 1.72
0.30 0.30 0.27 0.25
measured at 1.0 GPa and 400 8C and at 0.6 GPa and 25 8C for the pyroxenite (r ¼ 3.41 g cm23), mafic granulite (r ¼ 3.02 g cm23), mafic granulite (r ¼ 2.88 g cm23), and orthopyroxene-bearing felsic gneiss (r ¼ 2.68 g cm23) are listed in Table 2. As an example, the results at 1.0 GPa and 400 8C are Vp ¼ 7.17 km s21, Vs ¼ 4.24 km s21, Vp/Vs ¼ 1.69, Poisson’s ratio (s) ¼ 0.23 for the pyroxenite (MI97012402); Vp ¼ 6.93 km s21, Vs ¼ 3.81 km s21, Vp/Vs ¼ 1.82, s ¼ 0.28 for the mafic granulite (MI97012710); Vp ¼ 6.88 km s21, Vs ¼ 3.72 km s21, Vp/Vs ¼ 1.85, s ¼ 0.29 for the mafic granulite (MI96122602); and Vp ¼ 6.17 km s21, Vs ¼ 3.59 km s21 Vp/Vs ¼ 1.72, s ¼ 0.24 for the orthopyroxene-bearing felsic gneiss (MI97012401). Figure 2 shows the relationship between elasticwave velocity and pressure for the mafic granulite
(SiO2 ¼ 52.2 wt%) at 25 8C. The data exhibit relatively lower pressure dependence of velocity (dVp/dP or dVs/dP) above 0.6 GPa than below 0.6 GPa. The strong pressure dependence of velocity at lower pressure conditions is probably attributed to progressive closure of microcracks and decreasing rock porosity. The data indicate that rock porosity is mostly closed up to 0.6 GPa. At pressures from 0.6 to 1.0 GPa, Vp and Vs slightly increase or remain nearly constant. P- and S-wave velocities decrease with increasing temperature at pressures of 0.4, 0.7 and 1.0 GPa in all runs (Figs 3 and 4). In the run for mafic granulite (MI97012710, SiO2 ¼ 52.2 wt%) at 0.4 GPa an increasing temperature results in a sudden decrease in P- and S-wave velocities. This indicates thermal cracking at relatively low
Fig. 2. Pressure dependence of P- and S-wave velocities in mafic granulite (SiO2 ¼ 52.2 wt%) at 25 8C. Both P- and S-wave velocities strongly depend on pressure at relatively lower pressure (,0.6 GPa). At higher pressures from 0.6 to 1.0 GPa, P- and S-wave velocities slightly increase or remain nearly constant.
ELASTIC PROPERTIES OF META-IGNEOUS ROCKS
187
Fig. 3. P-wave velocities as a function of temperature at 0.4, 0.7 and 1.0 GPa for the four UHT meta-igneous rocks. Dashed lines are obtained by a least-squares fit. P-wave velocity decreases with increasing temperature at pressures of 0.4, 0.7 and 1.0 GPa in all runs.
pressure. Therefore the Vp and Vs values and corresponding elastic parameters (Vp/Vs and Poisson’s ratio) measured at higher pressure conditions (0.6–1.0 GPa) are considered to be the intrinsic properties of the UHT meta-igneous rocks at lower crustal depths.
P- and S-wave velocities and corresponding Vp/Vs and Poisson’s ratios measured at 1.0 GPa and 400 8C and at 0.6 GPa and 25 8C are listed in Table 2 for all rock samples. The velocity data represent intrinsic values of the P-wave and S-wave velocities of rocks without foliation
188
M. ISHIKAWA ET AL.
Fig. 4. S-wave velocities as a function of temperature at 0.4, 0.7 and 1.0 GPa for the four UHT meta-igneous rocks. Dashed lines are obtained by a least-squares fit. S-wave velocity decreases with increasing temperature at pressures of 0.4, 0.7 and 1.0 GPa in all runs. In the run for mafic granulite (SiO2 ¼ 52.2 wt%) at 0.4 GPa increase in temperature causes a relatively sudden decrease in Vp and Vs. Thermal cracking at relatively low pressures may cause this behaviour.
and lineation. The P- and S-wave velocities range from 6.17 km s21 (orthopyroxene-bearing felsic gneiss; SiO2 ¼ 65.4 wt%, r ¼ 2.68 g cm23) to 7.17 km s21 (pyroxenite; SiO2 ¼ 44.2 wt%, r ¼ 3.41 g cm23) and 2.34–4.93 km s21. The calculated P-wave acoustic impedance (rVp) of the
pyroxenite, mafic granulites, and felsic gneiss are 21.8 106 kg m22 s21, 26.0 106 kg m22 s21, 6 22 21 20.8 10 kg m s and 17.5 106 kg m22 s21, respectively. The reflection coefficient R ¼ (r2Vp2 2 r1Vp1)/(r2Vp2 þ r1Vp1), where r1Vp1 and r2Vp2 are the acoustic impedance of the first and
ELASTIC PROPERTIES OF META-IGNEOUS ROCKS
second medium, respectively. The values of the mafic granulites and the felsic gneiss range from 0.109 to 0.086 at 1.0 GPa and 25 8C. Figure 5 shows the velocity –density relationship of various metamorphosed igneous rocks from the Napier Complex, which were metamorphosed to UHT conditions. To compare velocities of the UHT meta-igneous rocks with minor amounts of retrograde biotite and those of meta-igneous rocks of granulite to amphibolite facies with relatively high amounts of biotite (Lu¨tzow-Holm Complex), the P- wave velocity data sets of Kitamura et al. (2001) are also shown in Figure 5. Both data sets show velocities that generally increase with densities in nearly linear manner. The comparison of both velocity –density data, however, shows that the UHT meta-igneous rocks have apparently higher velocities, basically as a result of their lower contents of hydrous
189
minerals such as hornblende and, in particular, biotite. The present results demonstrate that mineral assemblage reflecting both chemical composition and metamorphic grade is the most significant parameter influencing elastic-wave velocities. High modal contents of orthopyroxene and clinopyroxene are responsible for the relatively higher P-wave velocities and densities. The relatively low velocities of the metamorphosed igneous rocks from the Lu¨tzow-Holm Complex are due to higher abundance of hornblende and biotite. Figure 6 shows the measured Vp values for the Napier UHT metamorphosed igneous rocks at 0.6 GPa and 25 8C. The dashed line is obtained by fitting the measured Vp as a function of the SiO2 component of rock samples using the least-squares fitting method. We obtained a linear correlation expressed by Vp (km s21) ¼ 9.8469 2 0.0513SiO2 (wt%). The formula (continuous line) obtained
Fig. 5. Positive correlation between Vp and densities at 1.0 GPa. Filled symbols indicate the present results for the UHT meta-igneous rocks at 25 8C ( ) and 400 8C (B). The open symbols refer to results obtained for meta-igneous rocks from the Lu¨tzow-Holm Complex (after Kitamura et al. 2001). Velocity and density for minerals are also shown: Bt, biotite; Qtz, quartz; Ab, albite; An, anorthite; Hb, hornblende; Cpx, clinopyroxene; Opx, orthopyroxene (after Birch 1961; Anderson et al. 1968; Simmons & Wang 1971). LC, velocity and density of the lower crust beneath the Mizuho Plateau.
†
190
M. ISHIKAWA ET AL.
Fig. 6. P-wave velocities as a function of SiO2 wt% at 0.6 GPa and 25 8C. Filled symbols indicate the present results for the UHT meta-igneous rocks at 25 8C (†) and 400 8C (B). The open symbols refer to results obtained for metaigneous rocks from the Lu¨tzow-Holm Complex at 25 8C (W) and 400 8C (A) (after Kitamura et al. 2001). Dashed line (Vp ¼ 2 0.051 SiO2 þ 9.85) is a least-squares fit. Continuous line is the formula obtained for granulite-facies rocks of various regions after Rudnick & Fountain (1995).
from laboratory measurements of granulite-facies rocks (Rudnick & Fountain 1995) is also shown in this figure for comparison. The measured Vp values for the Napier UHT metamorphosed igneous rocks are significantly higher than the line from Rudnick & Fountain (1995). This difference could be attributed to the extremely ‘dry’ nature of the Napier UHT metamorphosed igneous rocks. The samples measured in this study have an extremely low abundance of hydrous minerals (less than 2.5 vol.% biotite; no amphibole). In the Vp –SiO2 diagram, the laboratory measurements of metamorphosed igneous rocks from the Lu¨tzow-Holm Complex (Kitamura et al. 2001) are also shown. The Vp values of rock samples of the Lu¨tzow-Holm Complex are generally lower than those of the Napier UHT metamorphosed igneous rocks, and the rock samples having relatively high abundance of biotite and/or hornblende are placed near or below the line from Rudnick & Fountain (1995).
Discussion In previous studies (Shingai et al. 2001; Ishikawa & Kanao 2002; Kanao & Ishikawa 2004; Kanao et al. 2004), the P-wave velocity structure of the Mizuho Plateau reported by Ikami et al. (1984) was interpreted as showing that the lower crust
(Vp ¼ 6.8–6.9 km s21) is composed of mafic granulites with a main mineral assemblage of orthopyroxene þ clinopyroxene þ plagioclase. Based on the recent seismic experiments of the SEAL project, however, Yoshii et al. (2004) proposed a new model of seismic structure of the Mizuho Plateau. In this model, the lower crust is characterized by Vp ¼ 6.56 km s21 and Vp/Vs ¼ 1.70. The Vp is about 0.3 km s21 lower than that reported by Ikami et al. (1984). Therefore, here we reconsider the crustal structure and its rock component of the lower crust beneath the Mizuho Plateau, and provide a clue to the new crustal model for this region. A collisional tectonic model has been proposed for the Lu¨tzow-Holm Complex in the Mizuho Plateau (Shiraishi et al. 1994, 1997), which suggests that the Lu¨tzow-Holm Complex collided with the Proterozoic Rayner Complex during the Pan-African orogeny and the Rayner Complex was amalgamated with the Napier Complex during Grenvillian time. Although the locations of Archaean Napier UHT metamorphosed igneous rocks are as distant as several hundred kilometres from the Mizuho Plateau, the possibility that the Napier Complex underlies the Lu¨tzow-Holm Complex in the Mizuho Plateau (Pan-African orogenic belt) cannot be entirely ruled out. The modern analogue of this tectonic model is that of
ELASTIC PROPERTIES OF META-IGNEOUS ROCKS
underthrusting of the Indian continental crust beneath the Himalayan orogenic belt. To petrologically interpret the lower crustal seismic velocity structure beneath the Mizuho Plateau (Yoshii et al. 2004), we calculated in situ P-wave velocity profile of the measured rock samples (orthopyroxene felsic gneiss, mafic granulites and pyroxenite) (Fig. 7). The calculation is made assuming a typical continental geothermal gradient. The comparison of the calculated velocity profiles of Napier UHT rocks with the seismic profile of Yoshii et al. (2004) suggests that the P-wave velocities and Vp/Vs of the Napier UHT rocks are very different from those of the seismic profile of Yoshii et al. (2004). Even if we assume that the lower crust is a mixture of mafic granulites and felsic gneisses, Vp and Vp/Vs cannot be explained simultaneously. Geophysical investigations at Mizuho Plateau including seismic and gravity surveys have revealed that the lower crust of this region has density of 2.93 g cm23 and velocity of 6.56 km s21 (Yoshii et al. 2004); both values are significantly lower than those obtained for the Napier rock samples. As shown in the velocity vs. density plots of Fig. 5, the lower crustal velocity of 6.56 km s21 and the density of 2.93 g cm23 suggested by Yoshii et al. (2004) cannot be explained by rocks having an anhydrous mineral assemblage
191
of orthopyroxene þ clinopyroxene þ plagioclase (gabbro, norite, or mafic granulite). Yoshii et al. (2004) suggested that felsic to intermediate granulites are the most probable lithological constituents of the middle and lower crust of this region. However, the density of felsic or intermediate rocks is significantly lower than that of the lower crust defined by seismic study, although the velocities of felsic or intermediate rocks are comparable with that of the lower crust (6.56 km s21). Biotite has a Vp –density relation that is unique among commonly occurring rock-forming minerals in high-grade metamorphosed lower crustal rocks; notably, low P-wave velocity and moderate density (Fig. 5). We propose that the velocity – density plot of biotite is central to the interpretation of the crustal structure. In particular, its P-wave velocity is significantly lower than those of metamorphosed igneous rocks from the Napier Complex. As in Fig. 5, higher abundance of biotite shifts P-wave velocities downward. The effect of biotite on P-wave velocity allows us to interpret that the low Vp and moderate density of the lower crust defined by Yoshii et al. (2004) may indicate relatively high abundance of biotite in the mafic lower crust beneath the Mizuho Plateau. According to rock velocity data, the silica composition of continental crust can be estimated using the formula Vp ¼ 8.91 2 0.038 SiO2 (Rudnick
Fig. 7. Comparison of rock velocities (continuous lines) with the seismic velocity structure (dashed lines) of the Mizuho Plateau (Yoshii et al. 2004). The P- and S- wave velocity profiles for the four Napier meta-igneous rocks were calculated by assuming a typical continental geotherm of Turcotte & Schubert (1982). The techniques used here have been described in previous studies (Shingai et al. 2001; Ishikawa & Kanao 2002). The S-wave velocity of the lower crust is calculated from Vp/Vs ¼ 1.70 of Yoshii et al. (2004).
192
M. ISHIKAWA ET AL.
& Fountain 1995). However, the velocity values measured in the metamorphosed igneous rocks from the Napier Complex and Lu¨tzow-Holm Complex differ slightly from those defined by this formula (Fig. 6). The velocity values of mafic granulites (sample MI96122602: SiO2 ¼ 52.2 wt% and Opx25Cpx20Pl53Opq3; sample MI96122710: SiO2 ¼ 49.5 wt% and Opx17Cpx14Pl62Bt1Opq7), pyroxenite (sample MI96122402: SiO2 ¼ 44.2 wt% and Opx52Cpx32Bt3Pl4Opq6) and orthopyroxene felsic gneiss (sample MI96122401: SiO2 ¼ 65.4 wt% and Opx7Cpx8Pl82Qtz3Opq2) are placed just above the line defined by this formula. The reason for the higher velocities of the Napier samples is because of the extremely dry nature of these rock samples. Mafic minerals of these rock samples, which were exposed to UHT metamorphism, are almost anhydrous. Therefore, the formula of Rudnick & Fountain (1995) is not suitable for estimating silica composition of anhydrous lower crusts. It has been reported that seismic velocities of biotite-bearing metamorphosed igneous rocks of the Lu¨tzow-Holm Complex are lower than predicted by the formula of Rudnick & Fountain (1995). The lower velocity of biotite-bearing metamorphosed igneous rocks is principally due to the higher modal abundance of biotite in these rocks Therefore, it is important to determine if the estimation of silica composition using the formula is applicable or not. In fact, as indicated in the Vp – density diagram (Fig. 5), for the eastern parts of Dronning Maud Land, biotite is considered to be one of the main mineral constituents of the lower crustal rocks. Accordingly, we note that the formula is not applicable to estimates of silica content of the eastern part of Dronning Maud Land. Although Yoshii et al. (2004) concluded that the lower crust region having a Vp of 6.56 km s21 is not mafic in composition, it is composed of biotitebearing mafic rocks. Therefore, we propose that the lower crust of the eastern part of Dronning Maud Land (Mizuho Plateau) is composed of biotite-bearing mafic rocks. Our interpretation is largely dependent on the seismic P-wave velocities and density from Yoshii et al. (2004). However, a few questions still remain about the adequacy and precision of these values. As noted by Yoshii et al. (2004), there are some misfits between observed and calculated gravity anomalies. In addition, there presumably exists a trade-off between density and layer thickness. This implies that there is ambiguity in the modelling of the crustal density, and the density model of Yoshii et al. (2004) is one of the most plausible candidates. Although these questions still remain at present, we have assumed that the seismic P-wave velocities and density from Yoshii et al. (2004) are of the best candidates at present. Therefore our hypothesis
that the low Vp and moderate density of the lower crust reflect high content of biotite in mafic lower crust remains a viable one. The biotite-bearing mafic lower crust could be a product of metasomatism involving an external fluid source. The Napier Complex is regarded as a fragment of Archaean continent (Sheraton et al. 1987). Zircon U –Pb dating of felsic gneisses in the Napier Complex shows a wide range of ages ranging from 3900 to 2400 Ma, reflecting a multistage growth history (Black et al. 1986; Harley & Black 1997; Suzuki et al. 2006). The Napier Complex is composed of metamorphosed igneous and sedimentary rocks, and its main rock constituent is orthopyroxene-bearing felsic gneiss of Archaean tonalite–trondhjemite– granodiorite (TTG) affinity (Ishikawa et al. 2000). The Napier Complex is characterized by UHT granulite-facies metamorphism where the metamorphic temperature exceeded 1000 8C (Harley & Hensen 1990). The Napier metamorphic rocks are also characterized by their extremely ‘dry’ nature, and it has been proposed that H2O was completely released by dehydration of precursors or partial melting prior to UHT metamorphism. The measured Vp values for the Napier mafic orthopyroxene–clinopyroxene– plagioclase granulite samples do not coincide with the lower crustal P-wave velocity and velocity ratio (Vp/Vs) suggested for the Mizuho Plateau. This suggests that the lower crustal rocks beneath the Mizuho Plateau are not characterized by extremely an ‘dry’ nature. In contrast to the Napier Complex, the neighbouring late Proterozoic Rayner Complex (Sheraton et al. 1987) and early Palaeozoic (550 –530 Ma) Lu¨tzow-Holm Complex (Hiroi et al. 1991; Shiraishi et al. 1989a, b) are predominantly composed of amphibolite- to granulite-facies rocks characterized by relatively high amounts of hydrous minerals such as hornblende and biotite (e.g. Motoyoshi et al. 1989; Motoyoshi & Ishikawa 1997). Some mafic rocks, in particular, contain a large amount of biotite, which reflects a metasomatic alternation process. Therefore, we infer that metasomatism of the lower crust is associated with that of the Lu¨tzow-Holm Complex. However, questions remain as to what is tectonic implication for biotiteforming metasomatism of the lower crust.
Conclusions For UHT meta-igneous rocks from the Napier Complex, we measured Vp and Vs at pressures up to 1.0 GPa and temperatures up to 400 8C using a piston-cylinder-type high-pressure apparatus. Vp and Vs increase with density and modal content of pyroxenes. The results at 1.0 GPa and 400 8C are Vp ¼ 7.17 km s21, Vs ¼ 4.24 km s21, Vp/Vs ¼ 1.69,
ELASTIC PROPERTIES OF META-IGNEOUS ROCKS
Poisson’s ratio (s) ¼ 0.23 for pyroxenite (SiO2 ¼ 44.2 wt%, density (r) ¼ 3.41 g cm21); Vp ¼ 6.93 km s21, Vs ¼ 3.81 km s21, Vp/Vs ¼ 1.82, s ¼ 0.28 for mafic granulite (SiO2 ¼ 52.2 wt%, r ¼ 3.02 g cm23); Vp ¼ 6.88 km s21, Vs ¼ 3.72 km s21, Vp/ Vs ¼ 1.85, s ¼ 0.29 for mafic granulite (SiO2 ¼ 49.5 wt%, r ¼ 2.88 g cm23); and Vp ¼ 6.17 km s21, Vs ¼ 3.59 km s21, Vp/Vs ¼ 1.72, s ¼ 0.24 for orthopyroxene felsic gneiss (SiO2 ¼ 65.4 wt%, r ¼ 2.68 g cm23), respectively. The comparison with data for biotite- and hornblende-bearing meta-igneous rocks from the Lu¨tzow-Holm Complex (Kitamura et al. 2001) reveals that the UHT meta-igneous rocks have significantly higher P-wave velocities. The relation between Vp, density and SiO2 indicates that the mineral assemblage, which reflects bulk chemical composition and metamorphic grade, is the most significant parameter influencing elastic properties. Vp –density and Vp –SiO2 diagrams show that a relatively large modal proportion of biotite is needed to explain the seismic low P-wave velocity (6.56 km s21) and density (2.93 g cm23) at lower crustal depth of the Mizuho Plateau. In conclusion, the lower crust of eastern Dronning Maud Land is suggested to be basic (mafic) in composition. Although the origin of the external fluid source is still obscure, the seismic P-wave velocity and density model suggests biotite-forming metasomatism acting in the mafic lower crust beneath the Mizuho Plateau during or after the Pan-African orogeny. We thank the reviewers K. Ito, M. Kanao and H. J. Mueller, and the editor M. Satish-Kumar for their constructive comments and suggestions. K. Shiraishi provided XRF facilities and K. Seno supported the bulkchemical analysis. Financial support to M.I. (19540502) and to M.A. (19340160) was provided by a Grant-in-Aid for Scientific Research from the Japanese Society for the Promotion of Science.
References A NDERSON , O. L., S CHREIBER , E., L IEBERMANN , R. C. & S OGA , N. 1968. Some elastic constant data on minerals relevant to geophysics. Review of Geophysics, 6, 491– 524. B IRCH , F. 1961. The velocity of compressional waves in rocks to 10 kbar, 2. Journal of Geophysical Research, 66, 2199– 2224. B LACK , L. P., W ILLIAMS , I. S. & C OMPSTON , W. 1986. Four zircon ages from one rock: The history of a 3930-Ma-old granulite from Mount Sones, Enderby Land, Antarctica. Contributions to Mineralogy and Petrology, 94, 427– 437. C HRISTENSEN , N. I. & M OONEY , W. D. 1995. Seismic velocity structure and composition of the continental crust: A global view. Journal of Geophysical Research, 100, 9761–9788. H ARLEY , S. & B LACK , L. P. 1997. A revised Archean chronology for the Napier Complex, Enderby Land,
193
from SHRIMP ion-microprobe studies. Antarctic Science, 9, 74– 91. H ARLEY , S. & H ENSEN , B. J. 1990. Archean and Proterozoic high-grade terranes of East Antarctica (40–808E). A case study of diversity in granulite facies metamorphism. In: A SHWORTH , J. R. & B ROWN , M. (eds) High Temperature Metamorphism and Crustal Anatexis. Unwin Hyman, London, 320– 370. H IROI , Y., S HIRAISHI , K. & M OTOYOSHI , Y. 1991. Late Proterozoic paired metamorphic complexes in East Antarctica, with special reference to the tectonic significance of ultramafic rocks. In: T HOMSON , M. R. A., C RAME , J. A. & T HOMSON , J. W. (eds) Geological Evolution of Antarctica. Cambridge University Press, Cambridge, 83– 87. H OKADA , T., M ISAWA , K., S HIRAISHI , K. & S UZUKI , S. 2003. Mid to late Archaean (3.3– 2.5 Ga) tonalitic crustal formation and high-grade metamorphism at Mt. Riiser-Larsen, Napier Complex, East Antarctica. Precambrian Research, 127, 215–228. I KAMI , A., I TO , K., S HIBUYA , K. & K AMINUMA , K. 1984. Deep crustal structure along the profile between Syowa and Mizuho Stations, East Antarctica. Memoirs of National Institute of Polar Research, Series C (Earth Science), 15, 69–89. I SHIKAWA , M. & K ANAO , M. 2002. Structure and collision tectonics of Pan-African orogenic belt— Scientific significance of the geotransect for a supercontinent: Gondwanaland. Bulletin of the Earthquake Research Institute, Tokyo University, 77, 287– 302. I SHIKAWA , M., H OKADA , T., I SHIZUKA , H., M IURA , H., S UZUKI , A., T AKADA , M. & Z WARTZ , D. P. 2000. Antarctic geological map series sheet 37 Mount Riiser-Larsen. National Institute Polar Research, Tokyo, 1 –23. I SHIZUKA , H., I SHIKAWA , M., H OKADA , T. & S UZUKI , S. 1998. Geology of Mount Riiser-Larsen of the Napier Complex, Enderby Land, East Antarctica. Polar Geoscience, 11, 154–171. I TO , K. & I KAMI , A. 1984. Upper crustal structure of Prince Olav Coast, East Antarctica. Memoirs of National Institute of Polar Research, Series C (Earth Science), 15, 13–18. I TO , K. & I KAMI , A. 1986. Crustal structure of the Mizuho Plateau, East Antarctica from geophysical data. Journal of Geodynamics, 6, 285– 296. I TO , K. & K ANAO , M. 1995. Detection of reflected waves from the lower crust on the Mizuho Plateau, East Antarctica. Antarctic Record, 39, 233 –242. K ANAO , M. & I SHIKAWA , M. 2004. Origins of the lower crustal reflectivity in the Lu¨tzow-Holm Complex, Enderby Land, East Antarctica. Earth, Planets, Space, 56, 151– 162. K ANAO , M., I SHIKAWA , M., Y AMASHITA , M., K AMINUMA , K. & B ROWN , L. D. 2004. Structure and evolution of the East Antarctic Lithosphere: Tectonic implications for the development and dispersal of Gondwana. Gondwana Research, 7, 31–41. K ITAMURA , K., I SHIKAWA , M., A RIMA , M. & S HIRAISHI , K. 2001. Laboratory measurements of P-wave velocity of granulites from Lu¨tzow-Holm Complex, East Antarctica: Preliminary report. Polar Geoscience, 14, 180– 194.
194
M. ISHIKAWA ET AL.
K ITAMURA , K., I SHIKAWA , M. & A RIMA , M. 2003. Petrological model of the northern Izu–Bonin– Mariana arc crust: constraints from high-pressure measurements of elastic wave velocities of the Tanzawa plutonic rocks, central Japan. Tectonophysics, 371, 213–221. K ONO , Y., I SHIKAWA , M. & A RIMA , M. 2004. Discontinuous change in temperature derivative of Vp in lower crustal rocks. Geophysical Research Letters, 31, L22601, doi:10.1029/2004GL020964. K ONO , Y., I SHIKAWA , M. & A RIMA , M. 2006. Laboratory measurements of P- and S-wave velocities in polycrystalline plagioclase and gabbronorite up to 700 8C and 1 GPa: Implications for the low velocity anomaly in the lower crust. Geophysical Research Letters, 33, L15314, doi:10.1029/ 2006GL026526. K ONO , Y., I SHIKAWA , M. & A RIMA , M. 2008. Effect of H2O released by dehydration of serpentine and chlorite on compressional wave velocities of peridotites at 1 GPa and up to 1000 8C. Physics of the Earth and Planetary Interiors, 161, 215–223. M IYAMACHI , H., M URAKAMI , H., T SUTSUI , T., T ODA , S., M INTA , T. & Y ANAGISAWA , M. 2001. A seismic refraction experiment in 2000 on the Mizuho Plateau, East Antarctica (JARE-41)—Outline of observations. Antarctic Record, 45, 101– 147. M OTOYOSHI , Y., M ATSUBARA , S. & M ATSUEDA , H. 1989. P– T evolution of the granulite-facies of the Lu¨tzow-Holm Bay region, East Antarctica. In: D ALY , J. S., C LIFF , R. A. & Y ARDLEY , B. W. D. (eds) Evolution of Metamorphic Belts. Geological Society, London, Special Publications, 43, 325 –329. M OTOYOSHI , Y. & I SHIKAWA , M. 1997. Metamorphic and structural evolution of granulites from Rundva˚gshetta, Lu¨tzow-Holm Bay, East Antarctica. In: R ICCI , C. A. (ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antartica, Siena, 65–72. M UELLER , H. J. 1995. Modelling the lower crust by simulation of the in situ conditions: an example from the Saxonian Erzgebirge. Physics of the Earth and Planetary Interiors, 92, 3 –15. M UELLER , H. J. & M ASSONNE , H.-J. 2001. Experimental high pressure investigation of partial melting in natural rocks and their influence on Vp and Vs. Physics and Chemistry of the Earth, Part A: Solid Earth and Geodesy, 26, 325–332. N ISHIMOTO , S., I SHIKAWA , M., A RIMA , M. & Y OSHIDA , T. 2005. Laboratory measurement of P-wave velocity in crustal and upper mantle xenoliths from Ichino-megata, NE Japan: ultrabasic hydrous lower crust beneath the NE Honshu arc. Tectonophysics, 396, 245– 259. R UDNICK , R. L. & F OUNTAIN , D. M. 1995. Nature and composition of the continental crust: A lower crustal perspective. Review of Geophysics, 33, 267– 309. S HERATON , J. W., T INGEY , R. J., B LACK , L. P., O FFE , L. A. & E LLIS , D. J. 1987. Geology of Enderby Land and
Western Kemp Land, Antarctica. Bulletin—Australian Bureau of Mineral Resources, Geology and Geophysics, 223, 1– 51. S HINGAI , E., I SHIKAWA , M. & A RIMA , M. 2001. P-wave velocity in ultrahigh temperature granulites from the Archean Napier Complex, East Antarctica. Polar Geoscience, 14, 165– 179. S HIRAISHI , K., H IROI , Y. & M OTOYOSHI , Y. 1989a. Antarctic Geological Map Series Sheet 12: Lu¨tzowHolm Bay. National Institute of Polar Research, Tokyo. S HIRAISHI , K., H IROI , Y. & M OTOYOSHI , Y. 1989b. Antarctic Geological Map Series Sheet 13: Prince Olav Coast. National Institute of Polar Research, Tokyo. S HIRAISHI , K., E LLIS , D. J., H IROI , Y., F ANNING , C. M., M OTOYOSHI , Y. & N AKAI , Y. 1994. Cambrian orogenic belt in East Antarctica and Sri Lanka: implication for Gondwana assembly. Journal of Geology, 102, 47–65. S HIRAISHI , K., E LLIS , D. J., F ANNING , C. M., H IROI , Y., KAGAMI, H. & M OTOYOSHI , Y. 1997. Re-examination of the metamorphic and protolith ages of the Rayner Complex, Antarctica: evidence for the Cambrian (Pan-African) regional metamorphic event. In: R ICCI , C. A. (ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antartica, Siena, 79–88. S IMMONS , G. & W ANG , H. 1971. Single Crystal Elastic Constants and Calculated Aggregate Properties: A Handbook. MIT Press, Cambridge, MA. S UZUKI , S., A RIMA , M., W ILLIAMS , I. S., S HIRAISHI , K. & K AGAMI , H. 2006. Thermal history of UHT metamorphism in the Napier Complex, East Antarctica: insights from zircon, monazite, and garnet ages. Journal of Geology, 114, 65–84. T SUTSUI , T., M URAKAMI , H., M IYAMACHI , H., T ODA , S. & K ANAO , M. 2001a. P-wave velocity structure of the ice sheet and the shallow crust beneath the Mizuho traverse route, East Antarctica, from seismic refraction analysis. Polar Geoscience, 14, 195–211. T SUTSUI , T., Y AMASHITA , M., M URAKAMI , H., M IYAMACHI , H., T ODA , S. & K ANAO , M. 2001b. Reflection profiling and velocity structure beneath the Mizuho traverse route, East Antarctica. Polar Geoscience, 14, 212– 225. T URCOTTE , D. L. & S CHUBERT , G. 1982. Geodynamics Applications of Continuum Physics to Geological Problem. Wiley, New York. Y OSHII , K., I TO , K., M IYAMACHI , H. & K ANAO , M. 2004. Crustal structure derived from seismic refractions and wide-angle reflections in the Mizuho Plateau, East Antarctica. Polar Geoscience, 17, 112–138. Y UKUTAKE , H. & I TO , K. 1984. Velocities of P and S waves for drilling core rocks at Syowa station, Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 33, 17– 27.
Early to middle Proterozoic dykes in the Mt. Riiser-Larsen area of the Napier Complex, East Antarctica: tectonic implications as deduced from geochemical studies SATOKO SUZUKI1,2, HIDEO ISHIZUKA3 & HIROO KAGAMI1 1
Graduate School of Science and Technology, Niigata University, Niigata, 950-2181, Japan 2
Present address: 20-8-2A, Matuba-cyo, Tokorozawa 359-0044, Japan (e-mail:
[email protected]) 3
Department of Geology, Kochi University, Kochi 780-8520, Japan
Abstract: NE–SW- and north–south-striking dykes were emplaced into ultrahigh-temperature (UHT) granulites apparently after UHT metamorphism in the Mt. Riiser-Larsen area of the Archaean Napier Complex, East Antarctica, of which the north–south-striking dykes interrupt the NE–SW-striking ones. The NE– SW-striking dykes are tholeiite basalt (THB) and highmagnesian andesite (HMA) in composition. The THB dykes display relict doleritic textures, whereas the HMA dykes shows blastoporphyritic textures characterized by phenocrysts of clinopyroxene and plagioclase. Both sets of dykes exhibit large ion lithophile element and light rare earth element enrichment and negative anomalies of Nb, Ti and/or P in a spider diagram normalized to primitive mantle, which is reminiscent of modern subduction-related arc volcanism or continental flood volcanism. The isotope ratios of the THB dykes define isochron ages of 2.0– 1.9 Ga: 1979 + 80 Ma in the Rb– Sr system (initial ratio (I0): 0.70239 + 0.00035) and 2078 + 104 Ma in the Sm– Nd system (I0: 0.50964 + 0.00012). Such moderate 87Sr/86Sr and low 143Nd/144Nd initial ratios may represent source materials closely related to the mantle wedge of a subduction zone. The north–south-striking dykes are compositionally divided into two basalt types. One is an alkaline basalt (AL) showing intergranular texture and characterized by high concentrations of incompatible elements, similar to those of ocean island basalt. They yield an isochron age of c. 1.2 Ga: 1161 + 238 Ma in the Rb –Sr system (I0: 0.7047 + 0.0012). The other type (THBm) is doleritic (ophitic) in texture, and has a tholeiitic affinity with a flat chondrite-normalized REE pattern, which is comparable with that of enriched mid-ocean ridge basalt. A comparison with dykes reported from other areas of the Napier Complex suggests that the north– south-striking dykes occur in restricted areas, whereas the NE–SW-striking dykes are more regional in occurrence. The 2.0–1.9 Ga magmatism of the NE– SW-striking dykes may have been related to the formation of continental crust of the Rayner Complex.
Proterozoic dykes of basic to intermediate compositions dominantly occur in the Precambrian East Antarctic Shield; for instance, in the Napier Complex of Enderby Land (Sheraton et al. 1980; Sheraton & Black 1981; Sheraton et al. 1987a, b), the southern Prince Charles Mountains of Mac Robertson Land (Tingey 1982), the Vestfold Hills of Princess Elizabeth Land (e.g. Oliver et al. 1982; Collerson & Sheraton 1986; Black et al. 1991) and the Bunger Hills of Queen Mary Land (Sheraton et al. 1990). Their intrusive ages form three clusters at c. 2.4 Ga, 2.2–1.8 Ga and 1.4– 1.2 Ga (Sheraton et al. 1987a; Lanyon et al. 1993). Of these, the Napier Complex includes various types of intrusive rocks such as metamorphosed dykes (mafic granulite), metamorphosed mafic dykes retaining doleritic textures, unmetamorphosed mafic dykes (dolerite), pegmatite, and lamproite, which occur in host ultrahigh-temperature (UHT)
metamorphic gneisses (e.g. Sheraton et al. 1987b; Owada et al. 2003). The geological record of the Napier Complex registers two episodes of tholeiitic dyke emplacement (Sheraton & Black 1981): early Proterozoic meta-tholeiites (2400 + 250 Ma) and middle Proterozoic tholeiites (1190 + 200 Ma). The suites have similar geochemical signatures such as large negative Nb and minor Ti anomalies. It is inferred that the early Proterozoic meta-tholeiite intrusions were related to the magmatic activity of unmetamorphosed high-magnesian (high-Mg) dykes, because they have similar isotopic signatures and their Rb–Sr isotope ratios co-define an isochron of 2350 + 48 Ma (Sheraton & Black 1981). Sheraton & Black (1981) considered that their source rocks, with almost uniform Sr isotopic composition, existed over a wide area in the Napier Complex, because of their homogeneous Sr initial ratios of 0.702. They then inferred that, if the
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 195 –210. DOI: 10.1144/SP.308.10 0305-8719/08/$15.00 # The Geological Society of London 2008.
196
S. SUZUKI ET AL.
source of the high-Mg suite was enriched in incompatible elements, the metasomatism must have occurred shortly before the generation of these dykes. Sheraton et al. (1987b) indicated that the early Proterozoic meta-tholeiites were emplaced at considerable depths (0.7–1.0 GPa) in the crust during the waning stages of granulite-facies metamorphism, based on calculation of equilibration pressures. Sandiford & Wilson (1983, 1984) described the metamorphosed dykes from the Fyfe Hills– Khmara Bay region, provisionally named the ‘Khmara Dykes’, and suggested that these correspond to the dykes of c. 2.4 Ga in other areas, based on their structural similarities. Sheraton & Black (1981) called the unmetamorphosed (dolerite) dykes in the Napier Complex the ‘Amundsen Dykes’. Uralitization observed in some ‘Amundsen Dykes’ has been regarded as a result of the late Proterozoic metamorphism, which has been related to metamorphism of the nearby Rayner Complex. The ‘Amundsen Dykes’ were divided into ‘Group-I tholeiites’ and ‘Group-II tholeiites’ by chemical and isotopic features. The ‘Group-I tholeiites’ define a c. 1.2 Ga age, but the age of the ‘Group-II tholeiites’ has been not determined because of their isotopic heterogeneity. Sheraton & Black (1981) indicated that the extensive Group-I tholeiites were derived from a uniform source region under Enderby Land during the middle Proterozoic. Black & James (1983) and Miyamoto et al. (2000) dated lamproite intrusions from Priestley Peak and Tonagh Island, respectively, at 0.5–0.4 Ga. In this paper, dyke rocks occurring in the Mt. Riiser-Larsen area are described; and for
some of these the Rb–Sr and Sm –Nd whole-rock isotopes are studied, and the igneous ages and tectonic settings are discussed in comparison with other dykes in the Napier Complex.
Basement rocks of the Mt. Riiser-Larsen area The Mt. Riiser-Larsen area in the western Napier Complex (Fig. 1) is underlain by metamorphosed Archaean basement rocks, which consist of an upper unit of orthogneisses with tonalitic –granodioritic compositions (Massive Gneiss Series) and a lower unit of layered paragneisses with quartzofeldspathic, siliceous or aluminous gneisses and orthogneisses having basaltic, A-type granitic or tonalitic–granodioritic composition (Layered Gneiss Series). The boundary between the units is transitional (Transitional Gneiss Series) (Ishizuka et al. 1998), in which metamorphosed anorthosite and ultramafic rocks occur as thin layers or blocks or pods (Fig. 2). Geochronological studies give protolith ages of 3.3–3.0 Ga and 2.6–2.5 Ga for tonalitic –granodioritic orthogneisses and c. 2.8 Ga and 2.6–2.4 Ga for granitic gneisses and paragneisses (zircon sensitive high-resolution ion microprobe (SHRIMP) ages by Hokada et al. (2003); zircon and monazite SHRIMP ages and garnet and pyroxene Sm –Nd internal isochron ages by Suzuki et al. (2006)). These basement rocks record UHT metamorphism at temperatures above 1100 8C (Motoyoshi & Hensen 1989; Hokada 2001; Ishizuka et al. 2002). The timing of peak UHT metamorphism has been recently considered
Fig. 1. Index map of the Napier Complex, Enderby Land, East Antarctica, showing the location of the Mt. RiiserLarsen area. The area of UHT metamorphism is outlined after Harley & Hensen (1990).
PROTEROZOIC DYKES OF NAPIER COMPLEX
197
Fig. 2. Simplified geological map of the Mt. Riiser-Larsen area (modified after Ishizuka et al. 1998), showing localities of samples analysed. The arrowhead tips and numerals in squares show the sampling points and sample numbers, respectively.
to be c. 2.59–2.55 Ga (zircon SHRIMP ages by Harley et al. (2001), Crowe et al. (2002) and Kelly & Harley (2005)) or c. 2.48–2.47 Ga (zircon SHRIMP ages by Carson et al. (2002), Hokada et al. (2003, 2004) and Hokada & Harley (2004); monazite ages based on the U–Th–total Pb chemical isochron method (CHIME; Suzuki & Adachi 1991) by Grew (1998), Asami et al. (2002) and Hokada & Motoyoshi (2006) zircon and monazite SHRIMP ages by Suzuki et al. (2006)). Ishikawa et al. (2000) documented two phases of folding in the Mt. Riiser-Larsen area: the first with axes aligned NNE– SSW and wavelengths up to 100 m and the forming a broad domal structure after the UHT conditions. They found that the development of deformation in this area is less pronounced than in the Casey Bay area (James & Black 1981; Black & James 1983; Sheraton et al. 1987b). Metamorphosed dykes, now represented by mafic granulite, occur parallel to the gneissosity in basement rocks, but some metamorphosed dykes are also observed to cross-cut the gneissosity (Ishizuka et al. 1998; Ishikawa et al. 2000). The latter metamorphosed dykes may correspond to the early Proterozoic meta-tholeiites described by Sheraton & Black (1981) and the ‘Khmara Dykes’
of Sandiford & Wilson (1983, 1984). A shear zone called the Riiser-Larsen Main Shear Zone (RLMSZ) with a maximum width of 400 m occurs in the western part of the area (Fig. 2), and consists of pseudotachylite rocks and mylonites derived from the surrounding host rocks. On the basis of mineral reaction textures after sapphirine and quartz, Hokada (1999) and Hokada et al. (2008) recognized different peak metamorphic conditions for the eastern part (.1040 8C and 0.6–0.8 GPa) and the western part (.1030 8C and 0.8–0.9 GPa) of the RLMSZ and suggested that the two parts were juxtaposed during subvertical movement of the RLMSZ.
Field occurrences and petrography of the dykes Throughout the Mt. Riiser-Larsen area, NE– SW-striking dykes occur abundantly, whereas in the western region north–south-striking dykes are also observed (Fig. 2). Most dyke rocks are black to dark grey in colour and massive, and range from several tens of centimetres to a few metres, rarely reaching 30 m, in width (Fig. 3). The following field observations indicate that the emplacement
198
S. SUZUKI ET AL.
Fig. 3. Field occurrences of the dykes in the Mt. Riiser-Larsen area. (a) An alkaline basalt dyke with north–south strike. Width of photograph is c. 80 m. (b) Tholeiite basalt dykes with NE–SW strike. Width of photograoh is c. 900 m.
of the dykes apparently postdated the Napier UHT metamorphism: (1) these dykes clearly cut the foliation of the host UHT metamorphic rocks; (2) a chilled margin (several centimetres in thickness) is commonly well developed at the contact with the host; (3) there is no sign of UHT metamorphic minerals and mineral assemblages in the dykes. No thermal effect is observed in the host rocks. The north–south-striking dykes interrupt the NE –SW-striking dykes, suggesting that the north–south-striking dykes apparently followed the NE–SW-striking dykes. Near the RLMSZ, the dyke rocks are also sheared to form mylonites or sometimes pseudotachylite rocks. The NE– SW-striking dykes have produced a vertical offset in the host gneisses for several tens of metres. The NE–SW-striking dykes are clearly offset by the north–south-striking dykes and the RLMSZ (Fig. 2). On the basis of relict mineral assemblages and whole-rock compositions as described below, the NE–SW-striking dykes are recognized as tholeiite basalt (THB) and high-magnesian andesite (HMA), and the north–south-striking ones as alkaline basalt (AL) and tholeiite basalt with enriched mid-ocean ridge basalt (E-MORB) affinity (THB-m). Exceptionally, one AL basalt dyke with a NE–SW strike occurs in the southeastern part of the area.
The THB dykes, which are most common in the area, include clinopyroxene and plagioclase (Fig. 4a) with minor magnetite, and commonly display relict doleritic (ophitic) textures in their central portion. The HMA dykes show blastoporphyritic textures composed of phenocrysts of clinopyroxene and plagioclase (Fig. 4b). The grain size of the HMA dykes is generally smaller than that of the THB dykes. The AL dykes retain magmatic textures in their central part, with phenocrysts of clinopyroxene, plagioclase and biotite embedded in a groundmass composed of plagioclase, clinopyroxene, biotite, K-feldspar, apatite and ilmenite (Fig. 4c). The AL dykes rarely exhibit strong mylonitization. The THB-m dykes are rarely found in this area, and consist of relict doleritic (ophitic) clinopyroxene and plagioclase (Fig. 4d). The minerals of all of the dyke rocks in the Mt. Riiser-Larsen area are altered to different extents, to blue–green hornblende, plagioclase, clinopyroxene, orthopyroxene, biotite, epidote, garnet, quartz, pyrite and chalcopyrite. These secondary minerals are characteristically fine-grained, and occur either partially replacing igneous clinopyroxene and plagioclase or scattered in the matrix. Most importantly, they show no sign of foliation. It is, therefore, likely that these secondary minerals were formed
PROTEROZOIC DYKES OF NAPIER COMPLEX
199
during the cooling stage after magmatic intrusion, that is, during so-called autometamorphism, and the fluid or water that was available for the formation of hydrous minerals such as hornblende and biotite may have been supplied from the same magma.
Analytical procedures
Fig. 4. Photomicrographs (crossed Nicols) of dykes in the Mt. Riiser-Larsen area. Scale bars represent 1 mm. (a) THB (tholeiitie basalt) dyke showing relict doleritic textures composed of clinopyroxene and plagioclase. Fine-grained minerals are metamorphic clinopyroxene. (b) HMA (high-magnesian andesite) dyke having blastoporphyritic textures composed of phenocrysts of clinopyroxene and plagioclase. (c) AL (alkaline basalt) dyke showing magmatic textures composed of clinopyroxene, plagioclase, biotite, K-feldspar and apatite. (d) THB-m (tholeiitie basalt with E-MORB composition) dyke showing relict doleritic textures composed of clinopyroxene and plagioclase. Cpx, clinopyroxene; Pl, plagioclase; m-Cpx, metamorphic clinopyroxene; Bt, biotite.
Rock samples of c. 20 cm3 volume were used for whole-rock analyses. Whole-rock chemical compositions were determined using fused glass discs by X-ray fluorescence (XRF) analysis at the National Institute of Polar Research in Japan, following the methods of Motoyoshi & Shiraishi (1995) and Motoyoshi et al. (1996). Some minor elements and rare earth elements (REE) were analysed by inductively coupled plasma mass spectrometry (ICP-MS) at Activation Laboratories Ltd, Canada. Table 1 shows representative whole-rock major and minor element, and REE analyses. Sr and Nd isotope analyses were performed at the Graduate School of Science and Technology, Niigata University. The whole-rock powder samples were dissolved with a HF þ HCl þ HNO3 þ HClO4 mixture in Teflon vials. The detailed dissolution process was based on that of Suzuki et al. (2006). The Rb–Sr and Sm–Nd whole-rock isotope compositions were measured by thermal ionization mass spectrometry (TIMS), using a MAT262 system equipped with nine dynamic Faraday cups. The 87Sr/86Sr and 143Nd/144Nd ratios were normalized to 86Sr/88 Sr ¼ 0.1194 and 146Nd/144Nd ¼ 0.7219, respectively. The normalized 87Sr/86Sr ratios were corrected using NBS987 standard of 87Sr/86Sr ¼ 0.710241. The 143Nd/144Nd ratios were corrected with JNdi-1 (Nd isotopic reference of Geological Survey of Japan) of 143Nd/ 144Nd ¼ 0.512115, which has been standardized by using the La Jolla international standard of 143Nd/144Nd ¼ 0.511858 (Tanaka et al. 2000). The averages of 87Sr/86Sr ratios of NBS987 and 143Nd/144Nd ratios of JNdi-1 during this study were 0.710257 + 0.000013 (2s, n ¼ 20) and 0.512072 + 0.000013 (2s, n ¼ 16), respectively. Total blanks for Sr and Nd were ,0.88 ng and ,0.39 ng, respectively. The 87Rb/86Sr and 147 Sm/144Nd ratios were calculated by using the concentrations of Rb, Sr, Sm and Nd in Table 1. Analytical errors for 87Rb/86Sr and 147Sm/144Nd ratios were 0.5% (1s) and 0.1% (1s), respectively. Details of column separation and isotope analyses were based on procedures of Miyazaki & Shuto (1998), Hamamoto et al. (2000) and Yuhara et al. (2000). The isochron ages and the initial isotope ratios were calculated using the software of Kawano (1994), which was based on the equation of York (1966) with the following decay constants: l 87 Rb ¼ 1.42 10211 a21 (Steiger & Ja¨ger 1977)
200
S. SUZUKI ET AL.
Table 1. Compositions of dykes from the Mt. Riiser-Larsen area, Napier Complex Sample no.: Dyke type: XRF SiO2 TiO2 Al2O3 Fe2O3 MnO MgO CaO Na2O K2O P2O5 Total
THB-5 HI97012207 tholeiite basalt
THB-7 HI97020701 tholeiite basalt
HMA-1 HI97012208 high-Mg andesite
AL-1 HI96123007 alkali basalt
THB-m-1 HI97020802 tholeiite basalt
49.77 0.85 15.42 12.32 0.18 6.16 11.13 2.03 0.41 0.06
51.78 1.89 12.61 16.97 0.21 4.39 8.49 2.39 1.22 0.25
55.19 1.57 11.16 14.09 0.16 4.92 7.71 3.08 1.44 0.14
44.86 3.24 14.70 15.71 0.21 5.40 8.58 2.80 1.80 0.96
49.46 1.37 13.16 14.48 0.20 6.86 10.88 2.06 0.16 0.08
98.33
Ba Co Cr Cu Nb Ni Rb Sr V Y Zn Zr
198.4 50.1 391.8 134.6 4.2 177.1 12.0 193.1 243.7 22.4 87.9 67.4
ICP-MS La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Ga Ge Sb Cs Hf Ta W Tl Pb Bi Th U
8.26 18.2 2.20 9.74 2.96 1.01 3.38 0.59 4.07 0.88 2.68 0.393 2.54 0.408 13 1.0 0.5 0.4 1.9 0.2 0.5 0.10 ,5 0.59 1.36 0.29
100.18 430.3 50.8 335.7 84.4 8.6 180.7 30.4 223.3 399.7 38.4 132.5 160.3 24.3 50.3 5.73 24.8 6.51 1.93 6.81 1.07 6.83 1.42 4.31 0.62 3.84 0.606 22 2.1 0.6 0.8 4.6 0.6 0.5 0.30 19 0.50 4.26 0.70
99.44 326.2 51.5 385.3 325.7 11.5 158.0 57.3 279.2 212.4 26.3 133.7 173.4 29.9 62.6 7.02 28.9 6.94 1.96 6.51 0.95 5.64 1.06 2.96 0.399 2.45 0.354 21 2.3 0.9 3.4 5.3 1.2 1.4 0.79 34 1.33 10.1 2.62
98.25 1242.6 49.2 196.0 44.0 39.8 94.2 46.1 423.3 239.6 29.5 122.2 157.1 44.2 83.2 8.98 36.2 7.55 2.82 6.89 0.95 5.54 1.09 3.08 0.406 2.59 0.396 22 2.3 0.5 0.9 4.1 2.7 1.0 0.41 16 1.61 4.82 0.84
98.71 37.1 54.0 415.8 236.0 5.1 203.8 3.3 124.4 356.2 26.4 97.6 81.4 5.20 12.9 1.74 8.64 2.93 1.06 3.60 0.64 4.15 0.87 2.60 0.376 2.30 0.366 14 2.4 0.5 0.3 2.1 0.3 ,0.2 0.06 ,5 0.29 0.60 0.14
PROTEROZOIC DYKES OF NAPIER COMPLEX
201
and l 147Sm ¼ 6.54 10212 a21 (Lugmair & Marti 1978). 1Sr and 1Nd values were calculated using the following Chondritic Uniform Reservoir values (¼ 0 Ma): 87Sr/86Sr ¼ 0.7045, 87Rb/86Sr ¼ 0.0827 (DePaolo 1988), 143Nd/144Nd ¼ 0.512638 and 147 Sm/144Nd ¼ 0.1966 (Goldstein et al. 1984).
Results and discussion Whole-rock compositions The THB dykes range from 49 to 52 wt% SiO2, of which those with 52 wt% SiO2 contents (THB-7 and -8) display mylonitic textures and micro-faults defined by fine-grained quartz and garnet. The SiO2 content shows no correlation wiht the FeO* (total iron)/MgO ratio, but the FeO* and TiO2 contents apparently increase with increasing FeO*/ MgO ratio (Figs 5 and 6a). This is indicative of tholeiitic differentiation, although some THB dykes
Fig. 5. AFM((Na2O þ K2O)–(FeO*)–(MgO)) diagram for the dykes in the Mt. Riiser-Larsen area. The curved line between typical tholeiite and calc-alkaline magma series is after Irvine & Baragar (1971).
Fig. 6. FeO*/MgO variation diagrams for (a) major and (b) minor elements of the dykes in the Mt. Riiser-Larsen area.
202
S. SUZUKI ET AL.
Fig. 6. Continued.
are altered to fine-grained blue– green hornblende, plagioclase, clinopyroxene, orthopyroxene, biotite, and epidote. The decreasing MgO, CaO and Al2O3 contents v. FeO*/MgO ratio are mainly due to clinopyroxene and plagioclase fractionations in the source magma, whereas the increasing TiO2, FeO*, MnO and V contents v. FeO*/MgO ratio suggest the beginning of magnetite fractionation. In a primitive mantle normalized trace element diagram, the THB dyke rocks are characterized by negative Nb, Sr, P and Ti anomalies (Fig. 7). The total large ion lithophile element (LILE) and REE contents increase with increasing FeO*/MgO ratios (Figs 7 and 8); this is also consistent with the fractionation model. However, the enriched LILE and REE contents of THB-7 and -8 may be caused by late-stage formation of mylonitic textures and micro-faults. As the chemical trends delineated by major and trace elements are not associated with the amount of secondary minerals such as hornblende, it is most likely that the whole-rock chemical compositions are almost free of any effect of secondary alteration.
The HMA dykes have SiO2 contents of about 55 wt%. The behaviour of whole-rock major and trace element compositions of the HMA dykes v. FeO*/ MgO ratio is inconsistent with that of THB dykes (Figs 5–8). The HMA dykes are characterized by high MgO (5 –6 wt%), Cr (380 –490 ppm), and Ni (160 –200 ppm) contents, and by negative anomalies of Nb, Sr, P and Ba in primitive mantle normalized trace element diagrams (Fig. 7). Heavy REE (HREE) concentrations of the HMA dykes are low as compared with the THB dykes (Fig. 8), in spite of the higher SiO2 concentrations of the HMA dykes. The depleted HREE and yttrium of the HMA dykes may suggest residual garnet in the source material. Hoek & Seitz (1995) proposed that the emplacement of a continental mafic dyke swarm reflects a combination of melt generation in the mantle and extension in the crust, and indicated that such magmatism is associated with active plate boundary settings. In such settings, dykes often show LILE enrichment with negative Nb, Sr, P, and Ti anomalies on a primitive mantle normalized trace
PROTEROZOIC DYKES OF NAPIER COMPLEX
Fig. 7. Primitive mantle normalized spider diagram of trace elements for the dykes in the Mt. Riiser-Larsen area. The normalizing values are after those of Sun & McDonough (1989).
Fig. 8. Chondrite-normalized REE patterns for the dykes in the Mt. Riiser-Larsen area. The normalizing values are after those of Sun & McDonough (1989).
203
204
S. SUZUKI ET AL.
element diagram, suggesting that basalts were derived from subcontinental lithospheric mantle: subduction-related arc basalts (e.g. Hall & Hughes 1987; Crawford et al. 1989; Tarney 1992) or continental flood basalts (e.g. Carlson 1991; Radhakrishna & Joseph 1998). The former option explains the LILE enrichment and trace element anomalies of the subduction-related arc basalts as being caused by addition of fluid released from a subducted slab to the overlying mantle wedge. On the other hand, many workers have reported continental flood basalts with subduction-related geochemical signatures of LILE enrichment often also showing negative Nb, Sr, P, and Ti anomalies (e.g. Gunn 2006). These ‘arc-like tholeiites’ have been considered to be derived from metasomatized lithospheric mantle with or without recycled components (e.g. Tarney 1992). As described above (Fig. 9), the sources of the THB and HMA dykes have subduction-related geochemical signatures, but it is still in dispute whether the tectonic setting available for the generation of the THB and HMA dykes is subduction-related arc volcanism or continental flood volcanism. Whole-rock chemical compositions of the AL dykes range from 44 to 45 wt% SiO2. These dykes have relatively high concentrations of incompatible elements such as P, Ti, Ba, Sr, Nb, and LREE, which are consistent with those of ocean island basalts (OIB). Figure 9 also indicates that the AL dykes originated from within-plate magma. The SiO2 contents of THB-m dykes range from 49 to 50 wt%. The mineral assemblage, texture, and whole-rock chemical compositions of the THB-m dykes are similar to those of the THB dykes. However, the behaviour of some elements clearly indicates incompatibility between the THB and
Fig. 9. The dyke rocks in the Mt. Riiser-Larsen area plotted in the Nb/Y v. Ti/Y basalt discrimination diagram. Fields for within-plate basalts, MORB, and volcanic-arc basalts (grey zone) are after Rollinson (1993; from Pearce 1982). Symbols are as in Figure 6.
the THB-m dykes (Figs 7 and 8). The THB-m dykes have lower K2O, Ba, Sr and Rb contents and higher TiO2 contents than the THB dykes. The THB-m dykes represent relatively flat patterns on primitive mantle normalized trace element and chondrite-normalized REE spiderdiagrams, which are rather similar to those of present-day EMORB (Fig. 8). The THB-m dykes are, however, characterized by slightly negative Nb and P anomalies on a primitive mantle normalized trace element diagram. The source magma of the THB-m dykes may be E-MORB-like magma developed in a back-arc basin.
Rb –Sr and Sm – Nd whole-rock isotope compositions Figure 10 displays the Rb–Sr and Sm –Nd isotopic compositions for the dykes (Table 2). The THB dykes yield isochron ages of 1979 + 80 Ma in the Rb–Sr system (I0: 0.70239 + 0.00035) and 2078 + 104 Ma in the Sm –Nd system (I0: 0.50964 + 0.00012). If we exclude samples THB-7 and -8, which are enriched in LILE and LREE, the calculated ages are 2042 + 130 Ma in
Fig. 10. (a) 87Rb/86Sr v. 87Sr/86Sr and (b) 147Sm/ 144Nd v. 143Nd/144Nd diagrams for whole-rock samples of the dykes in the Mt. Riiser-Larsen area.
Table 2. Rb, Sr, Sm and Nd concentrations and isotope data for the whole-rock samples of dykes in the Mt. Riiser-Larsen area Sample no.
AL AL-1 AL-2 AL-3 AL-4 THB-m THB-m-1 THB-m-2
87
Rb/86Sr
87
Sr/86Sr(2s) I0 Sr (T, Ga) Sm (ppm) Nd (ppm) (ICP-MS) (ICP-MS)
HI96122903 HI96123106 HI97010604 HI97011606 HI97012207 HI97012901 HI97020701 HI97020801
19.9 10.8 16.1 10.7 12.0 11.2 30.4 37.4
181 181 174 180 193 179 223 239
0.3438 0.1852 0.2538 0.1825 0.1859 0.1423 0.4455 0.5117
0.712183 (13) 0.707644 (14) 0.709897 (14) 0.707639 (14) 0.707541 (14) 0.706313 (14) 0.715346 (14) 0.716661 (12)
HI97012208 HI97020501
57.3 51.5
279 273
0.6825 0.6057
0.721215 (14) 0.720143 (13)
(T ¼ 2.0) 0.70185 0.70209 0.70229 0.70217 0.70197 0.70205 0.70199 0.70132 (T ¼ 2.0) 0.70076 0.70199
HI96123007 HI97010401 HI97010405 HI97021101
46.1 39.9 55.1 48.7
423 420 429 418
0.3282 0.2975 0.3975 0.3676
0.7101265 (14) 0.7097775 (13) 0.7114555 (13) 0.7107465 (12)
124 140
0.06978 0.703876 (14) 0.1011 0.704214 (14)
HI97020802 HI97020901
3.25 5.35
147
Sm/ Nd
144
143
Nd/144 Nd(2s)
1Nd (T, Ga)
4.88 3.36 3.68 3.82 2.96 2.29 6.51 8.15
16.3 10.9 12.2 13.0 9.74 7.52 24.8 31.4
0.1810 0.1863 0.1823 0.1776 0.1837 0.1841 0.1587 0.1569
6.94 6.31
28.9 25.6
0.1451 0.511674 (14) 0.1490 0.511703 (13)
(T ¼ 2.0) 25.9 26.0 25.6 – 26.3 26.0 25.8 26.0 (T ¼ 2.0) 25.2 25.2
(T ¼ 1.2) 0.70468 0.70484 0.70485 0.70464
7.55 8.08 8.67 8.33
36.2 38.5 41.9 40.0
0.1261 0.1269 0.1251 0.1259
(T ¼ 1.2) 25.3 24.8 24.7 24.7
– –
2.93 3.27
8.64 9.69
0.512118 (13) 0.512186 (18) 0.512147 (17) – 0.512133 (15) 0.512159 (14) 0.511812 (14) 0.511781 (14)
0.511832 (14) 0.511864 (14) 0.511854 (14) 0.511858 (14)
0.2050 0.5127035 (14) 0.2040 0.5126635 (14)
PROTEROZOIC DYKES OF NAPIER COMPLEX
NE –SW-striking dykes THB THB-1 THB-2 THB-3 THB-4 THB-5 THB-6 THB-7 THB-8 HMA HMA-1 HMA-2 North–South-striking dykes
Rb (ppm) Sr (ppm) (XRF) (XRF)
– –
205
206
S. SUZUKI ET AL.
the Rb –Sr system (I0: 0.7022 + 0.0004). These results suggest that the THB dykes intruded at about 2.0–1.9 Ga; this is the first report of such an age for dykes from the Napier Complex. The isotopic compositions of the HMA dykes plot close to the isochrons for the THB dykes. This result may be a coincidence, because the THB and HMA dykes originated from two different magmas. However, the THB and HMA dykes denote the NE–SW-striking dykes in this area, and hence may have intruded in a similar tectonic setting. The THB dykes have moderate to high 87 Sr/86Sr and low 143Nd/144Nd initial ratios: 1–8 for 1Sr (2.0 Ga) and 26 for 1Nd (2.0 Ga). These initial ratios may indicate that the source magmas originated from the mantle wedge in a subduction zone setting. The AL dykes yield an Rb – Sr whole-rock isochron age of 1161 + 238 Ma (I0: 0.7047+ 0.0012). The imprecision of the isochron age is mainly due to the narrow range in Rb/Sr of the AL dykes. In the Sm–Nd system, the AL dykes cannot define an isochron because of their homogeneous isotope ratios (Table 2). The Rb –Sr whole-rock isochron may represent the intrusion age of the AL dykes at c. 1.2 Ga. As the AL dykes were deformed by the RLMSZ, the development of the RLMSZ apparently postdated the emplacement of the AL dykes (at 1.2 Ga). The isotope ratios of the THB-m dykes are not comparable with those of the other dykes. The THB-m
dykes might be younger than the THB dykes, because of the field observation that the THB-m dykes cut the THB dykes. The ages of 2.0 –1.9 Ga for the THB and HMA dykes and c. 1.2 Ga for the AL dykes are comparable with the age clusters of other intrusive rocks in the Precambrian East Antarctic Shield (Sheraton et al. 1987a; Lanyon et al. 1993).
Comparison with other dykes in the Napier Complex Many dykes have been reported from the Napier Complex (Sheraton & Black 1981; Sheraton et al. 1987a, b). To compare these with the dykes in the Mt. Riiser-Larsen area, the relationship between the Nb/Zr and Zr/Y ratios of these dykes was examined (Fig. 11), because (1) Nb and Zr have similar bulk distribution coefficients in a basaltic system, and thus the Nb/Zr ratio does not change drastically during partial melting or fractional crystallization of basaltic magma, and (2) the Zr/Y ratio may increase as a result of the degree of fractionation in the same magma source. As shown in Figure 11, it is apparent that the AL dykes are very rare in other areas of the Napier Complex. Also, it has been shown that there is no report of dykes similar to THB-m from other areas. These facts suggest that the north–south-striking dykes, that is, the AL and THB-m dykes in a Mt. Riiser-Larsen
Fig. 11. Nb/Zr v. Zr/Y diagram for the dykes in the Mt. Riiser-Larsen area and the various dykes described by Sheraton et al. (1987b).
PROTEROZOIC DYKES OF NAPIER COMPLEX
area, may occur in a restricted area in the Napier Complex. In contrast, there are no dykes in the Mt. RiiserLarsen area that resemble the ‘Group-I tholeiites’ described by Sheraton & Black (1981) and Sheraton et al. (1987a, b) that intruded at c. 1.2 Ga (Fig. 11). The ‘Group-I tholeiites’ have relatively high Nb/Zr and Zr/Y ratios, and yield a Rb– Sr isochron age of 1190 + 250 Ma (I0: 0.7041 + 0.0005) (Sheraton & Black 1981). Sheraton & Black (1981) concluded that the ‘Group-I tholeiites’ were derived from a uniform source region under Enderby Land at about 1.2 Ga and extensively intruded into the Napier Complex, the western boundary of which was regarded as the eastern part of the Casey Bay area (Fig. 1). This implies that the Mt. Riiser-Larsen area has not been under the influence of the extensive magmatism of the ‘Group-I tholeiites’ at about 1.2 Ga. The THB and HMA dykes are chemically comparable with the c. 2.4 Ga meta-tholeiites and highMg dykes reported by Sheraton & Black (1981) and Sheraton et al. (1987a, b) (Fig. 11). The negative Nb anomaly, Nb/Zr ratios of 0.03–0.06, and Sr initial ratios of 0.702 also suggest this compatibility. However, the c. 2.4 Ga Sr isochron ages of metatholeiites and high-Mg dykes are clearly different from the c. 2.0–1.9 Ga age of the THB dykes. On the other hand, the THB and HMA dykes have similar compositional ranges to the ‘Group-II tholeiites’ of Sheraton & Black (1981) and Sheraton et al. (1987a, b), although some of the ‘Group-II tholeiites’ have the lower Nb/Zr (,0.04) and Zr/Y (,3.2) ratios (Fig. 11). It is also significant that some of ‘Group-II tholeiites’ have similar Rb–Sr isotope signatures to those of the THB and the HMA dykes (Sheraton & Black 1981). It is, therefore, most likely that the THB and HMA dykes may have been similar in origin to some ‘Group-II tholeiites’, although the Group-II tholeiites have been not dated because of their chemical and isotopical heterogeneity. The compositional range of the Nb/Zr and Zr/Y ratios of Qtz-tholeiites reported by Sheraton et al. (1987b) also overlaps that of the THB and the HMA dykes. These facts suggest that the THB and HMA dykes may be widely distributed in the Napier Complex. Asami et al. (2002) and Owada et al. (2003) reported a c. 2.0 Ga age for UHT granulites: monazite and zircon CHIME ages from several localities in the Amundsen Bay area and Sm–Nd garnet internal isochron ages from Tonagh Island, respectively. Asami et al. (2002) interpreted c. 2200 Ma and 2000 Ma CHIME ages on zoned monazites as discrete metamorphic events rather than events following UHT metamorphism at c. 2475 Ma. Owada et al. (2003) inferred that a tectonothermal event extensively affected the Napier Complex. The 2.0–1.9 Ga magmatism of the THB dykes may
207
have been related to the formation of continental crust of the Rayner Complex (Black et al. 1987; Kelly et al. 2002; Owada et al. 2003).
Summary (1) The dyke rocks were apparently emplaced after the Napier UHT metamorphism in the Mt. RiiserLarsen area, and are divided into four suites, based on field relations, mineralogy and geochemical features: tholeiite basalt (THB) and highmagnesian andesite (HMA) for the NE–SW-striking dykes, and alkaline basalt (AL) and tholeiite basalt (THB-m) for the north–south-striking dykes. The north–south-striking dykes interrupt the NE–SWstriking ones. (2) LILE and light REE (LREE) enrichment and negative anomalies of Nb, Ti and/or P in a spiderdiagram normalized to primitive mantle as observed in the NE–SW-striking dykes can be considered as analogues to subduction-related arc volcanism or continental flood volcanism. (3) The flat chondrite-normalized REE patterns and the slightly negative Nb and P anomalies for the THB-m dykes are suggestive of E-MORB origin. The high concentrations of incompatible elements such as P, Ti, Ba, Sr, Nb, and LREE in the AL dykes indicate their close affinity with OIB. (4) The Rb–Sr and Sm– Nd isotope data for the THB and the AL dykes define isochrons of 2.0– 1.9 Ga and 1.2 Ga, respectively, which are most likely to represent the ages of magmatism. (5) Comparison with other dykes in the Napier Complex indicates that the NE–SW-striking dykes occur extensively in the Napier Complex. The 2.0– 1.9 Ga magmatism of the NE–SW-striking dykes may have been related to the formation of continental crust of the Rayner Complex. We are grateful to T. Hokada and M. Ishikawa for their valuable advice, support in the fieldwork, and stimulating discussions. Thanks are also due to M. Arima, N. Tsuchiya, K. Suzuki, S. Wallis and R. Senda for their valuable suggestions. NIPR is thanked for supporting this study. The manuscript benefited greatly from detailed constructive reviews by K. Kullerud, M. Owada, C. J. Dobmeier and M. Satish-Kumar. The Japan Society for the Promotion of Science (JSPS) is acknowledged for financial support to S.S. while at the Graduate School of Science and Technology, Niigata University.
References A SAMI , M., S UZUKI , K. & G REW , E. S. 2002. Chemical Th–U –total Pb dating by electron microprobe analysis of monazite, xenotime and zircon from the Archean Napier Complex, East Antarctica: evidence for ultra-high-temperature metamorphism at 2400 Ma. Precambrian Research, 114, 249–275.
208
S. SUZUKI ET AL.
B LACK , L. P. & J AMES , P. R. 1983. Geological history of the Napier Complex of Enderby Land. In: O LIVER , R. L., J AMES , P. R. & J AGO , J. B. (eds) Antarctic Earth Science. Australian Academy of Science, Canberra, A.C.T. 11–15. B LACK , L. P., H ARLEY , S. L., S UN , S.-S. & M C C ULLOCH , M. T. 1987. The Rayner Complex of East Antarctica: complex isotopic systematics within a Proterozoic mobile belt. Journal of Metamorphic Geology, 5, 1 –26. B LACK , L. P., K INNY , P. D. & S HERATON , J. W. 1991. The difficulties of dating mafic dykes: an Antarctic example. Contributions to Mineralogy and Petrology, 109, 183–194. C ARLSON , R. W. 1991. Physical and chemical evidence on the cause and source characteristics of flood basalt volcanism. Australian Journal of Earth Sciences, 38, 525–544. C ARSON , C. J., A GUE , J. J. & C OATH , C. D. 2002. U–Pb geochronology from Tonagh Island, East Antarctica: implications for the timing of ultra-high temperature metamorphism of the Napier Complex. Precambrian Research, 116, 237–263. C OLLERSON , K. D. & S HERATON , J. W. 1986. Age and geochemical characteristics of a mafic dyke swarm in the Archaean Vestfold Block, Antarctica: Inferences about Proterozoic dyke emplacement in Gondwana. Journal of Petrology, 27, 853 –886. C RAWFORD , A. J., F ALLOON , T. J. & G REEN , D. H. 1989. Classification, petrogenesis and tectonic setting of boninites. In: C RAWFORD , A. J. (ed.) Boninites. Unwin Hyman, London, 2– 49. C ROWE , W. A., O SANAI , Y., T OYOSHIMA , T. & O WADA , M. 2002. SHRIMP geochronology of a mylonite zone on Tonagh Island: characterisation of the last highgrade tectonothermal event in the Napier Complex, East Antarctica. Polar Geoscience, 15, 17–36. D E P AOLO , D. J. 1988. Neodymium Isotope Geochemistry: an Introduction. Springer, New York. G OLDSTEIN , S. L., O’N IONS , R. K. & H AMILTON , P. J. 1984. A Sm– Nd study of atmospheric dusts and particulates from major river system. Earth and Planetary Science Letters, 70, 221–236. G REW , E. S. 1998. Boron and beryllium minerals in granulite-facies pegmatites for the origin and evolution of the Archean Napier Complex of East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 53, 74–93. G UNN , M. B. 2006. Geochemstry of igneous rocks. World Wide Web Address: http://www.geokem. com/index.html. H ALL , R. P. & H UGHES , D. J. 1987. Noritic dykes of southern West Greenland: early Proterozoic boninitic magmatism. Contributions to Mineralogy and Petrology, 97, 169–182. H AMAMOTO , T., Y UHARA , M., M IYAZAKI , T. ET AL . 2000. Rb, Sr, Sm and Nd separation from rocks, minerals and natural water using ion-exchange resin. Science Reports, Niigata University, Series E (Geology), 15, 49– 58. H ARLEY , S. L. & H ENSEN , B. J. 1990. Archaean and Proterozoic high-grade terranes of East Antarctica (40–808E): a case study of diversity in granulite facies metamorphism. In: A SHWORTH , J. R. & B ROWN , M.
(eds) High-temperature Metamorphism and Crustal Anatexis. Unwin Hyman, London, 320–370. H ARLEY , S. L., K ENNY , P. D., S NAPE , I. & B LACK , L. P. 2001. Zircon chemistry and the definition of events in Archaean granulite terrains. In: C ASSIDY , K. F., D UNPHY , J. M. & VAN K RANENDONK , M. J. (eds) 4th International Archaean Symposium—Abstracts. Geoscience Australia, Canberra, 511–513. H OEK , J. D. & S EITZ , H.-M. 1995. Continental mafic dyke swarms as tectonic indicators: as example from the Vestfold Hills, East Antarctica. Precambrian Research, 75, 121–139. H OKADA , T. 1999. Thermal evolution of the ultrahightemperature metamorphic rocks: in the Archaean Napier Complex, East Antarctica. PhD thesis, Graduate University for Advanced Studies, Tokyo. H OKADA , T. 2001. Feldspar thermometry in ultrahightemperature metamorphic rocks: evidence of crustal metamorphism attaining 1100 8C in the Archaean Napier Complex, East Antarctica. American Mineralogist, 86, 932 –938. H OKADA , T. & H ARLEY , S. L. 2004. Zircon growth in UHT leucosome: constraints from zircon–garnet rare earth elements (REE) relations in Napier Complex, East Antarctica. Journal of Mineralogical and Petrological Sciences, 99, 180–190. H OKADA , T. & M OTOYOSHI , Y. 2006. Electron microprobe technique for U– Th–Pb and REE chemistry of monazite, and its implications for pre-, peak- and postmetamorphic events of the Lu¨tzow-Holm Complex and the Napier Complex, East Antarctica. Polar Geoscience, 19, 118– 151. H OKADA , T., M ISAWA , K., S HIRAISHI , K., S UZUKI , S. & Y OKOYAMA , K. 2003. Mid to Late Archaean (3.3–2.5 Ga) tonalitic crustal formation and highgrade metamorphism at Mt. Riiser-Larsen, Napier Complex, East Antarctica. Precambrian Research, 127, 215–228. H OKADA , T., M ISAWA , K., Y OKOYAMA , K., S HIRAISHI , K. & Y AMAGUCHI , A. 2004. SHRIMP and electron microprobe chronology of UHT metamorphism in the Napier Complex, East Antarctica: implications for zircon growth at .1,000 8C. Contributions to Mineralogy and Petrology, 147, 1 –20. H OKADA , T., M OTOYOSHI , Y., S UZUKI , S., I SHIKAWA , M. & I SHIZUKA , H. 2008. Geodynamic evolution of Mt. Riiser-Larsen, Napier Complex, East Antarctica, with reference to UHT mineral associations and their reaction relations. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctic: a Key to the East–West Gondwana Connection. Geological Society, London, Special Publications, 308, 253–282. I RVINE , T. N. & B ARAGAR , W. R. A. 1971. A guide to the chemical classification of the common volcanic rocks. Canadian Journal of Earth Sciences, 8, 523– 548. I SHIKAWA , M., H OKADA , T., I SHIZUKA , H., M IURA , H., S UZUKI , S., T AKADA , M. & Z WARTZ , D. P. 2000. Geological map of Mt. Riiser-Larsen, Enderby Land, Antarctica. Antarctic Geological Map Series Sheet 37. National Institute of Polar Research, Tokyo. I SHIZUKA , H., I SHIKAWA , M., H OKADA , T. & S UZUKI , S. 1998. Geology of the Mt. Riiser-Larsen area of
PROTEROZOIC DYKES OF NAPIER COMPLEX the Napier Complex, Enderby Land, East Antarctica. Polar Geoscience, 11, 154– 171. I SHIZUKA , H., S UZUKI , S. & N AKAMURA , A. 2002. Peak temperatures of ultra-high temperature metamorphism of the Napier Complex, Enderby Land, East Antarctica, as deduced from porphyroclastic pyroxenes of meta-ultramafic rocks. Polar Geoscience, 15, 1–16. J AMES , P. R. & B LACK , L. P. 1981. A review of the structural evolution and geochronology of the Archaean Napier Complex of Enderby Land, Australian Antarctic Territory. In: GLOVER , J. E. & GROVES , D. I. (eds) Archaean Geology. Geological Society Australia Special Publication, 7, 71– 83. K AWANO , Y. 1994. Calculation program for isochron ages of Rb–Sr and Sm–Nd systems using personal computer. Geoinformatics, 5, 13– 19 [in Japanese with English abstract]. K ELLY , N. M. & H ARLEY , S. L. 2005. An integrated microtextural and chemical approach to zircon geochronology: refining the Archean history of the Napier Complex, East Antarctica. Contributions to Mineralogy and Petrology, 149, 57–84. K ELLY , N. M., C LARKE , G. L. & F ANNING , C. M. 2002. A two-stage evolution of the Neoproterozoic Rayner structural episode: new U– Pb sensitive high resolution ion microprobe constraints from the Oygarden Group, Rayner Complex, East Antarctica. Precambrian Research, 116, 307– 330. L ANYON , R., B LACK , L. P. & S EITZ , H.-M. 1993. U– Pb zircon dating of mafic dykes and its application to the Proterozoic geological history of the Vestfold Hills, East Antarctica. Contributions to Mineralogy and Petrology, 115, 184– 203. L UGMAIR , G. W. & M ARTI , K. 1978. Lunar initial 143 Nd/144Nd: differential evolution of the lunar crust and mantle. Earth and Planetary Science Letters, 39, 349–357. M IYAMOTO , T., G REW , E. S., S HERATON , J. W. ET AL . 2000. Lamproite dykes in the Napier Complex at Tonagh Island, Enderby Land, East Antarctica. Polar Geoscience, 13, 41–59. M IYAZAKI , T. & S HUTO , K. 1998. Sr and Nd isotope ratios of twelve GSJ rock reference samples. Geochemical Journal, 32, 345– 350. M OTOYOSHI , Y. & H ENSEN , B. J. 1989. Sapphirine– quartz–orthopyroxene symplectites after cordierite in the Archaean Napier Complex, Antarctica: evidence for a counterclockwise P –T path? European Journal of Mineralogy, 1, 467– 471. M OTOYOSHI , Y. & S HIRAISHI , K. 1995. Quantitative chemical analyses of rocks with X-ray fluorescence analyzer: (1) major elements. Antarctic Record, 39, 40–48. M OTOYOSHI , Y., I SHIZUKA , H. & S HIRAISHI , K. 1996. Quantitative chemical analyses of rocks with X-ray fluorescence analyzer: (2) trace elements. Antarctic Record, 40, 53– 63. O LIVER , R. L., J AMES , P. R., C OLLERSON , K. D. & R YAN , A. B. 1982. Precambrian geological relationships in the Vestfold Hills, Antarctica. In: C RADDOCK , C. (ed.) Antarctic Geoscience. University of Wisconsin Press, Madison, WI, 455– 464. O WADA , M., O SANAI , Y., T OYOSHIMA , T., T SUNOGAE , T., H OKADA , T., C ROWE , W. A. & K AGAMI , H. 2003.
209
Early Proterozoic tectonothermal events in the Napier Complex, East Antarctica: implications for the formation of East Gondwana. Gondwana Research, 6, 231– 240. P EARCE , J. A. 1982. Trace element characteristics of lavas from destructive plate boundaries. In: T HORPE , R. S. (ed.) Andesites. Wiley, Chichester, 525– 548. R ADHAKRISHNA , T. & J OSEPH , M. 1998. Geochemistry and petrogenesis of the Proterozoic dykes in Tamil Nadu, southern India: another example of the Archaean lithospheric mantle source. Geologische Rundschau, 87, 268– 282. R OLLINSON , H. 1993. Using Geochemical Data: Evaluation, Presentation, Interpretation. Longman, Singapore. S ANDIFORD , M. & W ILSON , C. J. L. 1983. The structural evolution of the Fyfe Hills– Khmara Bay region, Enderby Land, East Antarctica. Australian Journal of Earth Sciences, 31, 403 –426. S ANDIFORD , M. & W ILSON , C. J. L. 1984. The geology of the Fyfe Hills –Khmara Bay region. In: O LIVER , R. L., J AMES , P. R. & J AGO , J. B. (eds) Antarctic Earth Science. Australian Academy of Science, Canberra, A.C.T., 16–19. S HERATON , J. W. & B LACK , L. P. 1981. Geochemistry and geochronology of Proterozoic tholeiite dykes of East Antarctica: evidence for mantle metasomatism. Contributions to Mineralogy and Petrology, 78, 305– 317. S HERATON , J. W., O FFE , L. A., T INGEY , R. J. & E LLIS , D. J. 1980. Enderby Land, Antarctica—an unusual Precambrian high-grade metamorphic terrain. Journal of the Geological Society of Australia, 27, 1–18. S HERATON , J. W., T HOMSON , J. W. & C OLLERSON , K. D. 1987a. Mafic dyke swarms of Antarctica. In: H ALLS , H. C. & F AHRIG , W. F. (eds) Mafic Dyke Swarms. Geological Association of Canada, Special Papers, 34, 419–432. S HERATON , J. W., T INGEY , R. J., B LACK , L. P., O FFE , L. A. & E LLIS , D. J. 1987b. Geology of Enderby Land and western Kemp Land, Antarctica. Bulletin, Australian Bureau of Mineral Resources, 223, 1– 51. S HERATON , J. W., B LACK , L. P., M C C ULLOCH , M. T. & O LIVER , R. 1990. Age and origin of a compositionally varied mafic dyke swarm in the Bunger Hills, East Antarctica. Chemical Geology, 85, 215– 246. S TEIGER , R. H. & J A¨ GER , E. 1977. Subcommission on geochronology: convention on the use of decay constants in geo- and cosmochronology. Earth and Planetary Science Letters, 36, 359–362. S UN , S.-S. & M C D ONOUGH , W. F. 1989. Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In: S AUNDERS , A. D. & N ORRY , M. J. (eds) Magmatism in the Ocean Basins. Geological Society, London, Special Publications, 42, 313– 345. S UZUKI , K. & A DACHI , M. 1991. Precambrian provenance and Silurian metamorphism of the Tsubonosawa paragneiss in the South Kitakami terrane, Northeast Japan, revealed by the chemical Th–U– total Pb isochron ages of monazite, zircon and xenotime. Geochemical Journal, 25, 357–376.
210
S. SUZUKI ET AL.
S UZUKI , S., A RIMA , M., W ILLIAMS , I. S., S HIRAISHI , K. & K AGAMI , H. 2006. Thermal history of UHT metamorphism in the Napier Complex, East Antarctica: insights from SHRIMP zircon and monazite U–Pb ages and garnet Sm–Nd ages. Journal of Geology, 114, 65–84. T ANAKA , T., T OGASHI , S., K AMIOKA , H. ET AL . 2000. JNd-1: a neodymium isotopic reference in consistency with La Jolla neodymium. Chemical Geology, 168, 279– 281. T ARNEY , J. 1992. Geochemistry and significance of mafic dyke swarms in the Proterozoic. In: C ONDIE , K. C. (ed.) Proterozoic Crustal Evolution. Elsevier, Amsterdam, 151–179.
T INGEY , R. J. 1982. The geologic evolution of the Prince Charles Mountains—an Antarctic Archean cratonic blook. In: C RADDOCK , C. (ed.) Antarctic Geoscience. University of Wisconsin Press, Madison, WI, 455–464. Y ORK , D. 1966. Least-squares fitting of a straight line. Canadian Journal of Physics, 44, 1079–1086. Y UHARA , M., H AMAMOTO , T., K ONDO , H., I KAWA , T., K AGAMI , H. & S HUTO , K. 2000. Rb, Sr, Sm and Nd concentrations of GSJ, KIGAM and BCR-1 rock reference samples analyzed by isotope dilution method. Science Reports, Niigata University, Series E (Geology), 15, 23– 64.
Magmatic evolution and tectonic setting of metabasites from Lu¨tzow-Holm Complex, East Antarctica YOSHIMITSU SUDA1,2, YOSHINOBU KAWANO3, GREG YAXLEY4, HIROSHI KORENAGA5 & YOSHIKUNI HIROI5 1
National Institute of Polar Research, Kaga 1-chome, Itabashi-ku, Tokyo 173-8515, Japan 2
Present address: Center for Chronological Research, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8602, Japan (e-mail:
[email protected])
3
Center for Research and Development of Higher Education, Saga University, Honjo 1, Saga 840-8502, Japan 4
Research School of Earth Sciences, The Australian National University, Canberra, A.C.T. 0200, Australia
5
Department of Earth Sciences, Faculty of Science, Chiba University, Yayoi-cho 1-33, Chiba 260-8560, Japan Abstract: Metabasites from the Lu¨tzow-Holm Complex, East Antarctica, are the equivalent of metamorphosed ultramafic and mafic rocks with ultrabasic to intermediate compositions, which occur as layers and blocks in the quartzo-feldspathic or metasedimentary gneisses. Field occurrences and whole-rock geochemistry suggest that the ultramafic rocks are all cumulitic protoliths, whereas the mafic rocks are mostly basaltic protoliths including some cumulates. Moreover, in a regional context, the geochemistry of metabasites shifts from island arc to ocean-floor affinities in a southwesterly direction from the Prince Olav Coast to the Lu¨tzow-Holm Bay area. Neodymium isotopic data suggest that the metamorphic rocks from the Prince Olav Coast and the northern Lu¨tzow-Holm Bay areas were derived from immature continental crust formed by active Mesoproterozoic crustal growth, whereas those from the southern Lu¨tzow-Holm Bay area were derived from mature continental crust and oceanic crust of older age. Thus, these results suggest that the Lu¨tzow-Holm Complex includes lithological units with various origins and ages that were amalgamated by multiple subduction, and underwent high-grade metamorphism as a result of the final collision of East and West Gondwana during the Pan-African orogeny.
Many geological studies since the 1970s suggested that the Lu¨tzow-Holm Complex (LHC) of East Antarctica is a high-grade mobile belt with metamorphic ages concentrated in the time of the Pan-African orogeny between 520 and 560 Ma (e.g. Shiraishi et al. 1994, 2003). It has been concluded that the main regional metamorphism in the LHC must have been related to continent– continent collision between fragments of the Gondwana supercontinent, and that this region might include remnants of a possible suture between East and West Gondwana (e.g. Shiraishi et al. 1994; Kriegsman 1995; Fitzsimons 1996, 2000; Grunow et al. 1996; Wilson et al. 1997; Jacobs & Thomas 2004). Detailed mineralogical investigations have indicated that metamorphic rocks in the LHC are characterized by early formed kyanite that was replaced by sillimanite at peak temperature, and hence a medium-pressure type metamorphism with a clockwise style of P –T–t history is proposed
throughout the LHC (e.g. Hiroi et al. 1983, 1991). Metamorphic grade progressively increases across the complex, varying from amphibolite facies in the northeastern area through the amphibolite – granulite facies transition in the central area to granulite facies in the southwestern area (Fig. 1). The highest peak pressure and temperature conditions in the region have been estimated to be 9–11 kbar and 950–1040 8C at Rundva˚gshetta (Motoyoshi & Ishikawa 1997). Metamorphic lithologies of the LHC include pelitic to psammitic felsic gneiss, quartzofeldspathic gneiss, charnockite, mafic to intermediate rocks with subordinate khondalite, calcsilicates, marbles, granitic gneiss and ultramafic rocks. Post- and syn-metamorphic intrusions of amphibolite, basalt, andesite and granitic rocks are also present. Previous studies have revealed that the exposure of the ultramafic rocks is limited to the western part of the Prince Olav Coast and Lu¨tzow-Holm Bay regions, and that of the
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 211 –233. DOI: 10.1144/SP308.11 0305-8719/08/$15.00 # The Geological Society of London 2008.
212
Y. SUDA ET AL.
Fig. 1. Map of the Lu¨tzow-Holm Complex showing systematic increase in metamorphic grade from amphibolite facies to granulite facies along the Prince Olav Coast and the Lu¨tzow-Holm Bay regions after Hiroi et al. (1983) and Satish-Kumar et al. (2006a). The number of samples analysed in this study is shown in parenthesis (ultramafic rocks, mafic rocks).
khondalite–marble association is limited to the southern part of the Lu¨tzow-Holm Bay region (e.g. Hiroi et al. 1986; Satish-Kumar et al. 2006b). Moreover, Yoshida (1978) proposed three lithostratigraphic units for the metamorphic rocks of the Lu¨tzow-Holm Bay region: the Skallen Group, charnockite with quartzo-feldspathic gneiss and marble; the Ongul Group, charnockite with garnet –biotite gneiss; the Okuiwa Group, pink granite with biotite gneiss. These lithological varieties indicate that this complex is composed of different lithological units or tectonic blocks that have been amalgamated and that underwent highgrade metamorphism of Pan-African age. This study focuses on the metamorphosed mafic–ultramafic rocks that have been classified as metabasites by many previous workers (e.g. Yoshida 1978). Metamorphosed mafic rocks are generally considered to have igneous origins (e.g. Bucher & Frey 1994), and have been used as indicators to identify the source for metamorphic rocks from metamorphic terranes or complexes (e.g. Ulianov et al. 2006). This is because the tectonic environment of basalt, a typical basic igneous rock, is generally distinguishable on the basis of geochemistry (e.g. Pearce 2002). Hiroi et al. (1986) indicated that ultramafic rocks from the LHC originated from gabbroic rocks of tholeiitic affinities, whereas Kanisawa et al. (1987) suggested that the mafic rocks are of mid-ocean
ridge basalt (MORB) affinity. It is therefore plausible to assume that the metabasites in the LHC were derived from tectonically fractured fragments of oceanic crust of the pre-existing Mozambique Ocean that separated East and West Gondwana (e.g. Meert 2003). More recent work by Suda et al. (2006) indicated that layered metabasites, mainly mafic rocks, are metamorphosed equivalents of intrusive dykes or sills, which are associated with supracrustal rocks such as calc-silicates and quartzite preserving relics of volcanosedimentary lithological units. The mafic rocks with island arc affinities occur in the northeastern area of the LHC, whereas those with T-type (transitional) to E-type (enriched) MORB affinities occur in the central and southwestern areas in the LHC. The rocks of T-type MORB affinity have similar geochemical characteristics to basalts from an oceanic plateau or marginal sea basin. Although many workers have speculated on a possible precursor for the metabasites in the LHC, the effects of chemical alteration of protoliths during high-grade metamorphism and/or cooling have not been sufficiently evaluated. In this study, on the basis of evaluations of possible chemical alterations, we would like to re-examine the protoliths of metabasites in the LHC, and discuss their magmatic histories and tectonic environments, in addition to presenting new geochemical and isotopic data.
¨ TZOW-HOLM COMPLEX METABASITES FROM LU
Metabasites from the Lu¨tzow-Holm Complex Definitions Although ‘metabasites’ is a collective term for metamorphosed mafic rocks according to the Glossary of Geology (Jackson 1997), the observed metabasites in the LHC indicate mafic rocks and ultramafic rocks with ultrabasic through basic to intermediate compositions. Hiroi et al. (1986) confined ultramafic rocks to those in which CIPW normative M (olivine þ pyroxene þ ilmenite) is greater than 70%. On the other hand, Kanisawa et al. (1987) and Suda et al. (2006) defined mafic rocks as those with SiO2 contents between 45 and 52 wt%. In this study, we define ultramafic and mafic rocks on the basis of the following three factors: (1) SiO2 content; (2) colour index (i.e. modal abundance of opaque and mafic minerals); (3) normative M (hypersthene þ diopside þ olivine) content. Ultramafic rocks are defined as those that satisfy at least one condition from the following: SiO2 content ,46 wt%, colour index .90 and normative M .70. Ultramafic rock with a colour index ,45 is indicative of anorthitic rock (i.e. enriched in plagioclase but quartz-free). On the other hand, mafic rocks are defined as those with SiO2 content .46 wt%, colour index 50 –90 and normative M ,70. Quartzo-feldspathic rocks, such as hornblende gneiss and biotite gneiss, are distinguished from mafic rocks by a colour index ,50. Normative compositions were calculated based on an assumed Fe2O3/FeO ratio of 0.15 for metabasites of ultrabasic compositions, 0.2 for those of basic compositions and 0.3 for those of intermediate compositions, following the definitions of Middlemost (1989). Consequently, the SiO2 contents of the analysed ultramafic rocks and mafic rocks are 28– 55 wt% and 46–57 wt%, respectively. The normative M values of analysed ultramafic rocks and mafic rocks are 19 –97 and 25– 62, respectively. The colour indices of analysed ultramafic rocks and mafic rocks are 20– 100 and 50 –59, respectively.
Field occurrences Distributions of the metabasites analysed in this study are shown in Figure 1. The exposure area of the mafic rocks is distributed across the LHC, whereas that of the ultramafic rocks is limited to the transitional area and granulite-facies area. Previous studies indicated that the eastern end of the exposure area of the ultramafic rocks is Niban Rock (e.g. Hiroi et al. 1986). Figure 2 shows representative modes of occurrence of mafic and ultramafic rocks from the LHC. Mafic rocks in all
213
areas are characterized by a distinct foliation of hornblende and/or biotite, which is concordant with a gneissic fabric in the host rocks. Suda et al. (2006) indicated that the mode of occurrence of mafic rocks changes with increasing metamorphic grade (Fig. 1). Mafic rocks in the amphibolite-facies area occur as layers alternating with quartzofeldspathic gneiss and granitic gneiss. The layers vary from a few centimetres to several metres in thickness. Boundaries between mafic rocks and host quartzo-feldspathic rocks are blurred and irregular, which suggests that the mafic protolith was intruded into the quartzo-feldspathic rocks prior to metamorphism and deformation (Fig. 2a and b). Mafic rocks in the transitional area occur as layers alternating with quartzo-feldspathic gneiss and granitic gneiss, in which many open and tight folds are developed. The layers vary from decimetres to several metres in thickness. The mafic rocks in the granulite-facies area generally occur as layers, and sometimes as schlieren, lenses and blocks, which have thickness varying from several centimetres to metres and are hosted predominantly by charnockite. The mafic layers are often boudinaged, forming schlieric patches and flattened lenses. Ultramafic rocks in the LHC occur as rounded blocks, lenses and sometimes as layers, which have thickness varying from a few metres to several metres throughout the complex (Fig. 2c and d). The blocks of ultramafic rocks are generally composed of massive and polygonal mineral aggregates, and are characterized by some compositional layering. These layers vary from several centimetres to metres in thickness, and the direction of the layering is generally oblique to that of gneissic fabric in the host rocks (Fig. 2c). Figure 2d shows the mode of occurrence of ultramafic rocks at Akarui Point, in which the compositional layers vary from pyroxenitic through olivine-rich and garnet-rich layers to anorthositic layers. Anorthitic ultramafic rocks of Ongul Islands occur as layers, blocks and schlieren, which have thickness ranging from several metres to decimetres, and are hosted by leucocratic pyroxene gneiss with relatively blurred boundaries. Peak temperature conditions at around 1000 8C are estimated at Rundva˚gshetta. This temperature is high enough to cause dehydration melting of amphibolites (Johannes & Holtz 1996). Migmatitic structures are widely found in some metabasites in this area. These have centimetre-scale quartzofeldspathic pods and clots of pyroxenes suggesting in situ partial melting during peak metamorphism. In this study, we arbitrarily excluded samples that are possibly influenced by secondary alterations, such partial melting, interaction with host rocks and dykes, and weathering.
214
Y. SUDA ET AL.
Fig. 2. Outcrop photographs of the metabasites. (a) Irregular boundary between layered mafic rock (amphibolite) and quartzo-feldspathic rock (biotite gneiss) of Akebono Rock. (b) Boundary between layered mafic rock (amphibolite) and quartzo-feldspathic rock (biotite gneiss) of Akebono Rock, indicating that the mafic protolith has intruded into the quartzo-feldspathic protolith prior to metamorphism and deformation. (c) Boundary between a block of ultramafic rock and host quartzo-feldspathic rock (charnockite) at Rundva˚gshetta. Direction of the layerings developed in the ultramafic rock is discordant with the gneissosity of the host quartzo-feldspathic rock. (d) Compositional layerings developed in a block of ultramafic rock at Akarui Point, characterized by pyroxenitic, olivine-bearing, garnet-bearing and anorthositic layerings. Dotted lines in (c) and (d) indicate structural direction of the layerings.
Petrography Mafic rocks Mineral assemblages for representative mafic rocks (Fig. 3) are listed in Table 1. Detailed petrographic descriptions of the mafic rocks were presented by Suda et al. (2006), and are summarized as follows. The main constituent minerals are hornblende, clinopyroxene, orthopyroxene, biotite, garnet, plagioclase and quartz. Accessory minerals are apatite, epidote, titanite, chlorite and Fe–Ti oxides and sulphides. The dominant mineral assemblage varies from plagioclase þ hornblende (i.e. amphibolite) to clinopyroxene þ orthopyroxene þ plagioclase (i.e. mafic granulite), consistent with the three metamorphic areas defined in the LHC. Rarely, mafic granulites are found at Cape Hinode in the amphibolite-facies area (Hiroi et al. 2006; Suda et al. 2006). Aligned, millimetre-long grains of hornblende, biotite and pyroxene define a foliation. Clinopyroxene is occasionally replaced by
hornblende. Elsewhere, hornblende and pyroxene grains 0.5–1 mm across occur in a granoblastic fabric. Garnet porphyroblasts 2–5 mm across are commonly surrounded by symplectites of plagioclase þ orthopyroxene þ magnetite, and sometimes garnets are completely changed into a plagioclase þ orthopyroxene þ magnetite assemblage. This texture reflects the decompressional history of the LHC. The elongated shapes of coarse-grained biotite occur overprinted with the foliation defined by the elongated shapes of hornblendes and pyroxenes. This feature indicates that the biotite formed as a secondary mineral phase, suggesting secondary alteration such as potassium metasomatism after the formation of the peak metamorphic mineral assemblage.
Ultramafic rocks Detailed petrographic descriptions for the ultramafic rocks in the LHC have been presented by Hiroi
¨ TZOW-HOLM COMPLEX METABASITES FROM LU
215
Fig. 3. Photomicrographs of the mafic rocks (a) and ultramafic rocks (b– d) under plane-polarized light. (a) Hornblende granulite from Skallevikshalsen. (b) Ultramafic garnet–hornblende granulite from Skallevikshalsen. (c) Ultramafic spinel– garnet amphibolite from the Ongul Islands. (d) Ultramafic spinel– olivine granulite from Akarui Point.
et al. (1986), and are summarized as follows. The main constituent mineral assemblages of the ultramafic rocks vary from garnet þ orthopyroxene þ spinel þ plagioclase þ hornblende to olivine þ orthopyroxene þ spinel þ plagioclase. A garnet þ corundum mineral assemblage rarely occurs in the rocks from Akarui Point. Accessory minerals are spinel, sapphirine, gedrite, plagioclase, biotite, sillimanite, andalusite, kyanite, staurolite, carbonates (magnesite, calcite), chlorite, pyrite, pyrrhotite and apatite. The garnet grains exhibit porphyroblastic textures, are 2–30 mm across and contain finegrained corundum, sillimanite, sapphirine, plagioclase, biotite, quartz and rutile as inclusions. The quartz and rutile in the ultramafic rocks occur as inclusions in the garnet. The garnet is generally surrounded by symplectitic intergrowth of spinel þ orthopyroxene þ plagioclase or orthopyroxene þ plagioclase, and sometimes the porphyroblastic garnet is completely replaced by orthopyroxene þ plagioclase þ spinel assemblages. The cracks developed in the olivine and orthopyroxene are filled with biotites. Moreover, olivine is partially
replaced by serpentine, and orthopyroxene by chlorite. These features indicate retrogressive hydrothermal alteration and reactions.
Geochemistry Sample preparation and analytical techniques Whole-rock geochemical analysis was conducted on 53 samples of mafic rocks and 71 samples of ultramafic rocks. Major element and selected trace element abundances (Co, Cr, Cu, Nb, Ni, Rb, Sr, V, Y, Zn and Zr) for all samples were determined by X-ray fluorescence spectrometry (XRF) at the National Institute of Polar Research (RIX3000) and Analytical Research Center for Experimental Science of Saga University (JEOL60S7). Analytical protocols and conditions followed the techniques given by Motoyoshi & Shiraishi (1994) and Motoyoshi et al. (1996) for the RIX3000 and Kakubuchi et al. (1999) for the JEOL60S7. Precisions evaluated from the comparisons between
216
Table 1. Mineral assemblage for representative metabasites from the Lu¨tzow-Holm Complex Analysis no.
Locality
Area
Rock type
SU04122504A SU04122503B SU04122601A SU04121905 SU04122004 SU05010302 SU05010301E SU05010309B 95010706 95020409 95020410 95020501 95020503 95020506 SU05020702 94122504 SU05012501A SU05013009 SU05012806 95012705 SU05011807A SU05011809 SU05011901 SU05011904 SU05012003B SU05012004 SU05012006 SU05012013 95012307 95012603 SU05011001 SU05011003B SU05011004 SU05011005B
3 7 35 40 44 10 76 78 KY-22 KY-14 KY-15 KY-17 KY-19 KY-21 82 KY-11 87 89 86 KY-8 14 95 96 98 100 101 102 105 KY-5 KY-6 16 17 110 111
AR AR AR CH CH AP AP AP CO OI OI OI OI OI LA LA SVN SVN HO SKL SVH SVH SVH SVH SVH SVH SVH SVH RVK RVK RVH RVH RVH RVH
AF AF AF AF AF T T T T GF GF GF GF GF GF GF GF GF GF GF GF GF GF GF GF GF GF GF GF GF GF GF GF GF
amph. amph. amph. cpx amph. hbl gran. amph. cpx amph. cpx amph. hornblendite um. spl-grt amph. um. spl amph. um. grt-hbl gran. um. cpx amph. pyroxenite gran. um. grt amph. um. hbl gran. hbl gran. gran. hornblendite grt-hbl gran. hbl gran. hbl gran. hbl gran. um. hbl gran. um. hbl gran. amph. cpx amph. um. hbl gran. um. amph. um. hbl gran. um. gran. hbl gran. hbl gran.
Cpx
W W
Opx
4
4 W
Hbl
Grt
* * * * * * *
Bt
Spl
W tr tr tr
W *
4
W
* *
W
tr
W W W W W
4
W
W
W W
W
W W W
W W W
W W W W W
W W W W W
W W
W
W W W W
W W W W
W W 4
*
W W W * * * * *
W *
Qtz
Om
CI
M
DO
W W W W W W W W W
W W
tr 4
69 62 72 65 55 63 53 60 93 ,34 ,20 57 90 100 50 43 90 59 60 97 68 67 62 71 70 63 63 72 100 90 96 91 88 60
38 42 37 41 34 37 38 38 51 20 23 41 50 66 33 20 50 34 37 49 49 55 44 40 33 31 36 46 76 47 59 58 60 48
A A A D A A A A D C C E D E B B D A A D A E A A B B A A E B E E E E
* *
W 4
4 4 W tr tr
4 4 tr tr 4 W 4 4
tr
* *
W 4
4 tr
4 W W 4 W W W W W W W W W W W W W
tr tr tr tr 4 tr
W *
tr W W
*
4
tr tr
Pl
tr tr 4
4 tr
tr tr tr tr 4 tr tr tr
tr tr 4
*, .40%; W, 40 –10%; 4, 10 –3%; tr, , 3% in modal abundances. AR, Akebono Rock; CH, Cape Hinode; GR, Gobanme Rock; AP, Akarui Point; CO, Cape Omega; OI, Ongul Islands; LA, Langhovde; HO, Honno¨r Oku-iwa; SVN, Skarvsnes; SVH, Skallevikshalsen; RVH, Rundva˚gshetta; AF, amphibolite-facies area; T, transitional area; GF, granulite-facies area; um, ultramafic; amph., amphibolite; gran., granulite; cpx, clinopyroxene; opx, orthopyroxene; hbl, hornblende; grt, garnet; bt, biotite; spl, spinel; pl, plagioclase; qtz, quartz; om; opaque minerals; CI, colour index (modal abundance of mafic and opaque minerals); M, CIPW normative composition of mafic minerals (hypersthene þ diopside þ olivine); Do, compositional domains (A, B, C, D and E) defined in Figure 5. (See Fig. 1 for Locality and metamorphic areas (Area).)
Y. SUDA ET AL.
Sample no.
¨ TZOW-HOLM COMPLEX METABASITES FROM LU
recommended values after Imai et al. (1995) and analysed values for geochemical reference samples (GSJ Japan: JA-2, JB-1a, JB-3 and JGb-1) were ,5% for major elements, V, Zn, Rb, Sr and Zr, ,10% for Cr and Nb, and ,15% for Co, Ni and Y. Selective trace element compositions (Sc, Co, Ni, Ga, Rb, Sr, Y, Zr, Nb, Ba, Lu, Hf, Pb, Th and U) including rare earth element (REE) compositions for 24 samples of metabasites were determined by laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS) at the Research School of Earth Sciences, Australian National University (ANU) using fused glass discs. The analytical techniques followed the method described by Eggins (2003). We used 43Ca as the internal reference isotope for data reduction, based on the value previously obtained by XRF. Typical analytical precision (2s) for elements analysed by LA-ICP-MS was ,2% for Ti, Sr and Ba, 2–5% for Sc, V, Y, Zr, Nb, La and Ce, 5–10% for Rb, Nd, Eu, Gd, Er, Yb, Hf, Ta and Th, and 10–15% for Sm, Lu, Pb and U. Detection limits were less than 0.10 ppm for most analysed trace elements. Results of wholerock analyses by XRF and LA-ICP-MS are listed in Table 2. Sr and Nd isotopic ratios for 34 samples of metabasites, and Sm and Nd compositions for 13 samples of metabasites were determined by thermal ionization mass spectrometry (TIMS) (MAT-262) at Niigata University. Procedures for extraction of Sm and Nd from rock powders following those given by Kawano et al. (1999). 143Nd/144Nd ratios were normalized to 146Nd/144Nd ¼ 0.7219. Average 143 Nd/144Nd ratio of JNdi-1 (Standard of Geological Survey of Japan) during this analysis was 0.512070 + 0.000008 (2s, n ¼ 7) corresponding to 143 Nd/144Nd 0.511858 of La Jolla (Tanaka et al. 2000). The 87Sr/86Sr ratios were normalized to 86 Sr/88Sr 0.1194. The average 86Sr/88Sr ratio of NBS987 during this analysis was 0.710211 + 0.0000011 (2s, n ¼ 6). Results of analysis by mass spectrometry are listed in Table 3.
Chemical alteration Fundamental factors that control the mobility of elements in metamorphic rocks are the abundance of elements in the protolith, concentration of elements in fluid, distribution of coefficients between fluid and rocks, and distribution and stability of elements in the secondary minerals (e.g. Floyd & Winchester 1978). In most cases, under hydrated condition, large ion lithophile elements (LILE), such as Sr, Rb and K, are generally considered mobile, whereas incompatible high field strength elements (HFSE), such as Ti, Zr, Y, Nb and REE, and compatible elements, such as Cr, Ni and V, are generally considered immobile
217
(e.g. Floyd & Winchester 1978). In rare cases, the REE are mobilized by a carbonatization related to high-CO2 metamorphic fluid (Hynes 1980), weathering (Melfi et al. 1990) and hydrothermal alteration (Valsami & Cann 1992). Under microscopic observation, the occurrences of biotite grains in metabasites suggest that most of them formed after the formation of the peak metamorphic assemblage, and they may have formed as a result of alkaline metasomatism during cooling, as described above. To evaluate the mobility of elements during the formation of secondary biotite, diagrams in Figure 4 show the relations between concentrations of selected elements and modal abundances of biotite in the mafic rocks with basaltic protoliths (Domain A in Fig. 5). Concentrations of K2O show a good positive correlation with modal abundance of biotite. This element is generally compatible with biotite. Of the other elements, the concentrations of TiO2, Zr, Nb, REE and LaN/YbN (chondrite-normalized) roughly show positive correlations, especially as concentrations of these elements differ between the biotitefree samples and the biotite-bearing samples. TiO2, Nb and Zr are slightly compatible with biotite, but the REE are incompatible. Therefore, titanite and/ or apatite must have been the reservoirs for the REE. In contrast to these elements, the SiO2/ Al2O3 ratio and Mg-number do not exhibit any correlation, and they do not have behave differently between the biotite-free samples and the biotitebearing samples. Modal abundances of biotite in metabasites are not significantly correlated with the metamorphic grades defined in the LHC (Fig. 1). Therefore, the diagrams in Figure 4 suggest that not only the LILE, such as K2O, Sr and Rb, but also the HFSE, such as Ti, Nb, Zr and REE, must have been mobilized during formation of a secondary phase of biotite, which may result from the introduction of an alkaline fluid. On the other hand, the SiO2/Al2O3 ratio and Mg-number are unaffected and might have preserved their protolithic values. Carbonate minerals are found in some ultramafic rocks, suggesting a carbonate metasomatism. We also excluded such samples in the discussion on protoliths.
Major and trace elements The Mg-number v. SiO2/Al2O3 diagram (Kempton & Harmon 1992; Kempton et al. 1997) for metabasites in the LHC is shown in Figure 5. Generally, Mg-number decreases with increasing magmatic differentiation, whereas the SiO2/Al2O3 ratio reflects the magmatic fractionation series: alkaline, tholeiite and calc-alkaline. Moreover, Mg-number will decrease with accumulation of Fe–Ti oxides, and increase with accumulation of pyroxenes,
Table 2. Major oxides (wt%) and trace element abundances (ppm) for metabasites from the Lu¨tzow-Holm Complex
XRF analysis SiO2 TiO2 Al2O3 Fe2O3 T MnO MgO CaO Na2O K2 O P 2 O5
35 40 44 10 76 78 KY-22* KY-14* KY-15* KY-17* KY-19* KY-21* AR CH CH AP AP AP CO OI OI OI OI OI amph. cpx amph. hbl gran. amph. cpx amph. cpx amph. hornblendite um. spl–grt amph. um. spl amph. um. grt –hbl gran. um. cpx amph. pyroxenite
82 LA gran.
KY-11* 87 LA SVN um. grt amph. um. hbl gran.
51.30 1.01 14.53 15.32 0.26 4.80 8.72 3.85 0.13 0.11
51.69 0.79 15.29 11.55 0.19 7.28 11.00 2.84 0.24 0.08
48.73 0.81 15.67 11.57 0.28 8.76 8.83 3.22 1.95 0.04
47.55 0.45 18.16 7.63 0.17 9.29 14.34 2.25 0.62 0.07
51.40 0.64 16.49 9.71 0.17 6.85 10.35 3.83 0.87 0.07
50.61 1.00 15.64 11.12 0.15 7.47 9.80 3.31 0.62 0.12
50.76 1.03 15.23 10.61 0.20 7.45 10.58 3.28 0.82 0.11
50.40 1.23 14.69 10.93 0.21 6.93 11.14 3.53 0.55 0.24
45.30 1.22 12.28 12.03 0.16 14.02 12.29 1.26 0.22 0.25
45.24 0.42 25.68 7.79 0.09 4.84 12.85 2.03 1.01 0.01
43.47 0.21 23.73 11.89 0.34 6.15 10.99 1.76 1.00 0.02
44.81 0.12 15.96 12.29 0.17 12.80 9.23 2.16 1.57 0.01
42.47 1.47 13.84 12.96 0.14 12.51 12.30 1.45 2.15 0.17
49.09 0.32 6.79 9.22 0.15 17.59 11.80 0.64 3.49 0.00
46.78 3.79 14.91 15.11 0.15 6.26 8.72 1.52 1.51 0.75
40.51 1.56 23.70 10.77 0.09 2.11 20.75 0.17 0.18 0.19
46.13 0.54 13.73 10.63 0.15 17.56 8.79 1.52 0.55 0.13
100.04 42.2 12.5 56.7 5.9 6.3 1.3 1.2 3.5 233 467 18.1 97.4 38.8
100.94 59.5 76.7 55.2 129 57.5 0.1 48.5 2.1 185 304 18.9 76.2 35.6
99.85 63.8 279 42.5 204 n.d. 2.4 76.7 65.8 159 221 18.6 117 39.5
100.52 74.0 130 33.6 150 18.4 3.7 38.6 4.7 334 144 14.0 68.3 51.7
100.36 62.2 148 37.1 79.7 n.d. 2.6 29.2 3.7 171 226 14.3 63.6 39.3
99.83 61.1 378 41.1 234 31.0 6.6 83.9 8.7 233 211 32.8 131 92.2
100.06 62.1 175 40.9 256 53.6 3.4 70.2 8.3 222 230 26.1 83.3 111
99.85 59.7 212 40.4 172 52.9 5.5 53.7 12.4 335 257 29.1 86.8 92.7
99.02 73.1 60 48.3 964 24 6 354 4 215 203 26 74 109
99.98 59.2 176 28.0 87 3 4 7 10 1057 112 9 61 44
99.54 54.7 198 32.9 127 n.d. 3 8 11 482 59 10 101 23
99.12 70.8 311 37.9 44 3 4 7 30 381 26 5 151 16
99.47 69.2 437 35.2 257 8 5 26 46 174 205 25 201 61
99.09 81.7 556 32.3 2687 6 10 552 249 87 75 54 164 27
99.49 49.1 253 48.8 105 34.3 14.8 43.1 131 39.5 209 50.3 167 344
100.03 31.4 45 38.4 299 41 16 56 7 1735 219 54 42 443
99.71 79.4 87.7 61.3 1059 70.6 2.8 382 11.4 302 111 10.7 84.2 67.0
40.5 305 51.8 59.1 1.5 179 18.1 30.0 1.0 74.9 0.9 4.6 0.1 0.1 2.0 4.8 0.8 4.3 1.6 0.7 2.2 0.4 2.9 0.6 1.9 1.9 0.3
39.9 249 44.7 90 58.3 153 18.4 38.6 2.3 292 1.2 2.3 0.2 0.4 2.0 4.7 0.8 4.3 1.7 0.6 2.5 0.4 3.1 0.7 2.0 2.0 0.3
40.5 174 43.2 45.0 2.2 329 13.8 44.3 4.1 132 1.2 2.0 0.4 0.5 4.1 9.7 1.4 6.6 1.8 0.7 2.2 0.4 2.4 0.5 1.4 1.4 0.2
35.8 250 40.0 34.8 2.3 167 16.4 35.3 2.2 141 1.1 2.1 0.5 0.2 3.4 7.9 1.2 5.7 1.8 0.7 2.3 0.4 2.8 0.6 1.8 1.8 0.3
36.2 226 42.1 94 7.3 229 34.2 80.4 7.0 362 2.3 3.8 0.4 0.2 8.5 19.3 3.0 14.3 4.2 1.8 5.2 0.9 6.0 1.3 3.7 3.6 0.5
37.4 261 43.7 76.4 6.2 215 26.9 103 3.9 185 2.6 3.3 0.6 0.1 7.1 15.1 2.3 11.3 3.2 1.3 4.0 0.7 4.6 1.0 3.0 2.9 0.4
45.2 299 44.5 59.0 9.2 350 31.3 86.1 5.5 211 2.5 6.5 1.0 0.3 11.4 26.7 4.1 18.9 4.9 1.7 5.4 0.8 5.6 1.2 3.3 3.1 0.5
LA-ICP-MS analysis Sc 42.1 V 485 Co 68.2 Ni 7.0 Rb 2.3 Sr 237 Y 19.4 Zr 38.0 Nb 0.7 Ba 37.9 Hf 1.1 Pb 3.9 Th 0.3 U 0.1 La 3.3 Ce 7.5 Pr 1.2 Nd 6.1 Sm 2.0 Eu 0.8 Gd 2.7 Tb 0.4 Dy 3.2 Ho 0.7 Er 2.0 Yb 2.0 Lu 0.3
31.1 233 46.8 43.9 129 40.0 50.7 343 15.7 361 8.5 4.2 2.3 1.2 31.3 77.6 11.8 53.3 12.8 3.3 12.3 1.7 10.2 2.0 5.1 4.2 0.6
20.4 136 88.9 430 9.3 294 12.1 62.5 3.3 94 1.6 8.7 1.6 0.9 9.2 21.2 3.1 13.9 3.2 1.0 2.9 0.4 2.3 0.5 1.2 1.1 0.2
Y. SUDA ET AL.
Total Mg. no. Ba Co Cr Cu Nb Ni Rb Sr V Y Zn Zr
7 AR amph.
218
Analysis no.: 3 Location: AR Rock type: amph.
Table 2. Continued 86 HO gran.
KY-8* SKL hornblendite
14 SVH grt– hbl gran.
95 SVH hbl gran.
96 SVH hbl gran.
XRF analysis SiO2 TiO2 Al2O3 Fe2O3 T MnO MgO CaO Na2O K2O P2O5
48.14 1.97 15.19 11.98 0.15 7.98 8.82 2.73 2.55 0.28
48.31 1.78 15.95 13.19 0.18 7.36 9.47 1.90 1.17 0.36
41.23 3.21 12.96 15.01 0.19 10.92 12.05 1.86 1.69 0.45
46.57 1.18 13.33 15.59 0.25 7.71 13.35 1.52 0.42 0.09
47.73 1.19 11.70 14.93 0.24 9.35 12.93 1.34 0.31 0.09
48.90 1.28 14.12 12.31 0.18 7.86 12.19 2.13 0.80 0.11
Total 99.79 Mg-no. 60.8 Ba 233 Co 43.7 Cr 282 Cu 57.0 Nb 8.2 Ni 89.8 Rb 98.9 Sr 294 V 202 Y 29.6 Zn 132 Zr 146 LA-ICP-MS analysis Sc 31.0 V 218 Co 46.5 Ni 89 Rb 91.0 Sr 278 Y 29.5 Zr 138 Nb 8.1 Ba 259 Hf 3.7 Pb 6.2 Th 2.0 U 0.7 La 16.3 Ce 37.1 Pr 5.2 Nd 22.4 Sm 5.4 Eu 1.9 Gd 5.6 Tb 0.8 Dy 5.4 Ho 1.1 Er 3.1 Yb 3.0 Lu 0.5
99.69 56.5 200 50.3 133 26.9 8.4 52.2 20.9 269 206 34.9 109 189
99.59 62.9 667 44.0 132 18 14 68 28 370 294 42 184 234
100.01 53.6 10.1 53.1 101 111 4.8 48.1 0.7 88.4 360 26.3 98.1 55.8
99.79 59.4 n.d. 53.6 811 103 5.2 80.0 2.1 86.6 386 28.5 104 51.7
49.7 374 57.6 61.3 1.1 113 26.5 49.4 4.0 38.4 1.4 4.5 n.d. n.d. 2.9 8.2 1.4 7.3 2.6 1.0 3.6 0.6 4.3 0.9 2.8 2.7 0.4
65.3 433 60.4 96.0 1.1 85.5 30.7 47.8 4.9 18.8 1.6 4.0 0.3 0.2 3.5 9.7 1.6 8.1 2.7 1.0 3.7 0.7 5.0 1.2 3.5 3.6 0.6
32.8 229 55.7 59.9 17.7 261 36.2 182 8.3 245 4.7 4.6 1.0 0.3 15.5 36.7 5.4 24.7 6.2 1.9 6.7 1.0 6.6 1.4 3.8 3.5 0.5
98 SVH hbl gran.
100 SVH um. hbl gran.
101 SVH um. hbl gran.
102 SVH amph.
105 SVH cpx amph.
48.00 0.85 15.39 12.15 0.16 8.36 11.12 2.37 1.02 0.06
45.57 3.49 14.46 16.45 0.23 6.00 8.76 3.02 1.01 0.47
45.01 3.45 14.94 16.50 0.22 5.02 8.78 3.06 1.79 0.45
47.01 1.76 15.59 13.01 0.18 7.39 9.49 3.16 1.84 0.15
46.88 1.50 13.68 14.26 0.25 7.74 12.66 2.37 0.94 0.11
99.88 59.8 78.2 46.4 256 128 7.0 101 10.3 167 297 25.8 94.2 97.6
99.47 61.6 62.7 47.2 183 1021 4.3 99.9 11.9 150 228 15.9 113 46.2
99.46 46.0 182 53.9 108 116 15.0 60.2 15.0 298 265 44.1 136 250
99.20 41.5 182 50.6 73.8 93.7 11.2 42.2 59.0 267 277 48.9 164 224
99.58 57.0 176 46.3 129 31.4 5.0 89.0 68.0 285 247 24.7 137 95.0
100.41 55.9 24.2 46.2 109 88.0 6.6 66.2 16.7 191 363 26.1 139 78.0
42.8 345 52.4 100 8.0 160 27.1 87 7.1 101 2.6 9.4 0.4 0.2 8.3 19.7 2.9 13.2 3.7 1.3 4.4 0.7 4.9 1.1 2.9 2.7 0.4
36.3 244 54.6 104 8.6 142 17.0 42.1 4.1 77.3 1.3 8.0 1.4 1.0 8.5 18.2 2.4 9.3 2.4 1.0 2.8 0.4 3.0 0.7 1.8 1.7 0.3
32.3 301 53.3 73.8 14.5 290 46.3 255 15.9 275 6.5 9.0 2.2 0.6 20.1 46.7 6.9 31.3 8.0 2.6 8.7 1.3 8.6 1.8 4.9 4.5 0.7
34.5 328 46.9 52.1 53.9 261 48.8 224 13.8 266 5.9 19.0 4.0 1.7 29.3 63.7 8.5 35.5 8.5 2.6 9.0 1.4 9.0 1.9 5.1 4.7 0.7
33.3 270 49.5 96.8 61.6 278 25.4 86.9 5.7 204 2.6 12.3 3.1 1.2 19.7 32.8 3.6 14.0 3.6 1.5 4.2 0.7 4.6 1.0 2.7 2.5 0.4
47.3 415 47.6 79.0 16.2 185 27.0 72.9 6.2 64.4 2.2 7.2 1.4 0.5 7.0 16.0 2.5 11.9 3.5 1.3 4.4 0.7 4.9 1.1 2.9 2.7 0.4
KY-5* RVK um. hbl gran.
KY-6* RVK um. amph.
16 RVH um. hbl gran.
17 RVH um. gran.
45.21 0.43 5.01 15.81 0.17 28.18 4.71 0.70 0.70 0.07
41.43 2.07 13.07 17.83 0.15 9.72 10.95 1.37 2.47 0.62
48.56 1.06 7.16 14.09 0.20 15.60 7.61 1.17 3.50 0.21
49.72 1.00 7.47 13.79 0.22 14.65 7.65 1.53 3.10 0.19
49.92 0.42 9.09 12.00 0.21 16.22 9.16 1.90 1.04 0.04
50.99 1.40 10.39 15.28 0.26 8.55 8.86 3.48 0.79 0.28
100.99 80.6 76 80.8 2801 10 3 568 8 122 117 12 201 54
99.68 56.0 259 47.6 604 11 13 101 43 92 306 90 325 346
99.15 72.1 423 58.7 793 11.5 9.0 563 320 140 195 18.3 118 98.8
99.33 71.2 248 56.4 797 1.7 9.2 527 324 110 179 21.4 136 100
99.99 75.9 373 54.8 1405 3.2 5.3 207 32.5 127 169 10.5 98.7 27.0
100.29 56.6 102 60.3 460 37.8 17.8 166 9.4 373 277 40.8 134 98.8
19.2 215 72.6 637 295 135 14.6 98.3 9.5 394 2.6 4.0 0.9 0.6 19.8 45.8 6.0 23.7 4.6 1.1 3.9 0.5 3.0 0.6 1.5 1.2 0.2
18.1 199 67.3 594 305 106 17.6 96.4 10.7 237 2.6 5.0 0.9 1.1 24.3 55.8 7.4 27.9 5.3 0.9 4.3 0.6 3.2 0.6 1.6 1.4 0.2
33.2 193 70.9 259 29.9 120 11.2 25.2 4.8 383 0.9 5.0 0.8 0.5 14.4 33.0 4.2 15.2 2.8 0.9 2.3 0.3 2.0 0.4 1.1 1.1 0.2
28.3 302 69.2 188 8.2 389 42.6 97 18.5 127 3.0 11.6 1.7 0.5 25.6 66.8 10.1 43.1 10.2 1.6 9.1 1.3 8.1 1.6 4.4 4.0 0.6
*Samples analysed at Saga University (Co value at NIPR); others at NIPR and/or ANU. Fe2O3 T, total Fe as Fe2O3. n.d., not determined. Mg-number ¼ Mg/(Mg þ 0.85FeT). Abbreviations as in Table 1.
110 RVH hbl gran.
111 RVH hbl gran.
219
89 SVN hbl gran.
¨ TZOW-HOLM COMPLEX METABASITES FROM LU
Analysis no.: Location: Rock type:
220
Table 3. Rb–Sr and Sm –Nd isotopic data for metabasites from the Lu¨tzow-Holm Complex Analysis no.
Rb (ppm)
Sr (ppm)
AR AR AR CH CH AP AP AP CO OI OI OI OI OI LA LA SVN SVN HO SKL SVH SVH SVH SVH SVH SVH SVH SVH RVK RVK RVH RVH RVH RVH
3.5* 2.1* 65.8* 4.7* 3.7* 8.7* 8.3* 12.4* 4* 10* 11* 30* 46* 249* 131* 7* 11.4* 98.9* 20.9* 28* 0.7* 2.1* 10.3* 11.9* 15.0* 59.0* 68.0* 16.7* 8* 43* 320* 324* 32.5* 9.4*
233* 185* 159* 334* 171* 233* 222* 335* 215* 1057* 482* 381* 174* 87* 39.5* 1735* 302* 294* 269* 370* 88.4* 86.6* 167* 150* 298* 267* 285* 191* 122* 92* 140* 110* 127* 373*
87
Sr/86Sr + 2s
0.703316 + 11 0.703168 + 14 0.719471 + 14 0.703127 + 12 0.703696 + 14 0.704191 + 14 0.704064 + 13 0.704060 + 13 0.704740 + 12 0.703415 + 11 0.706235 + 14 0.705923 + 14 0.711161 + 11 0.771852 + 14 0.787446 + 14 0.708587 + 12 0.704616 + 14 0.712577 + 13 0.705855 + 14 0.710959 + 14 0.711976 + 14 0.711699 + 14 0.720393 + 14 0.729260 + 14 0.705832 + 11 0.716720 + 13 0.719552 + 14 0.717465 + 14 0.710959 + 12 0.719330 + 15 0.778563 + 14 0.800290 + 14 0.729156 + 13 0.720260 + 12 0.704080 + 08
87
Rb/86Sr Sm (ppm) 0.04 0.03 1.20 0.04 0.06 0.11 0.11 0.11 0.05 0.03 0.07 0.23 0.77 8.33 9.69 0.01 0.11 0.97 0.22 0.22 0.02 0.07 0.18 0.23 0.15 0.64 0.69 0.25 0.19 1.35 6.66 8.58 0.74 0.07
2.0† 1.6† 1.7† 1.8† 1.8† 4.2† 3.2† 4.9† 4.66‡ 2.05‡ 3.90‡ 2.09‡ 16.48‡ 9.10‡ 12.8† 11.67† 3.2† 5.4† 6.2† 5.83‡ 2.6† 2.7† 3.7† 2.4† 8.0† 8.5† 3.6† 3.5† 1.34‡ 21.31‡ 4.6† 5.3† 2.8† 10.2†
Nd (ppm) 6.1† 4.3† 4.3† 6.6† 5.7† 14.3† 11.3† 18.9† 15.58‡ 8.98‡ 25.05‡ 14.26‡ 64.20‡ 26.98‡ 53.3† 54.70† 13.9† 22.4† 24.7† 31.84‡ 7.3† 8.1† 13.2† 9.3† 31.3† 35.5† 14.0† 11.9† 5.01‡ 79.04‡ 23.7† 27.9† 15.2† 43.1†
143
Sm/144Nd + 2s
0.512764 + 14 0.513023 + 14 0.513139 + 14 0.512687 + 14 0.512775 + 14 0.512783 + 14 0.512723 + 41 0.512601 + 14 0.512620 + 13 0.512405 + 34 0.512201 + 14 0.512296 + 10 0.512442 + 12 0.512466 + 13 0.512377 + 14 0.512151 + 14 0.512426 + 16 0.512160 + 14 0.512501 + 14 0.512373 + 14 0.512840 + 14 0.512550 + 14 0.512203 + 14 0.512385 + 14 0.512244 + 14 0.512400 + 13 0.512384 + 14 0.512352 + 31 0.512306 + 14 0.512126 + 16 0.511362 + 14 0.511262 + 14 0.511560 + 14 0.511367 + 13 0.512737 + 14
147
Sm/144Nd 1Nd(550 Ma) 0.200 0.228 0.237 0.168 0.191 0.178 0.174 0.159 0.181 0.138 0.094 0.088 0.155 0.204 0.145 0.129 0.139 0.145 0.152 0.111 0.214 0.202 0.170 0.154 0.155 0.145 0.154 0.181 0.162 0.163 0.118 0.116 0.110 0.143
18.0 21.0 22.7 18.7 18.8 19.9 19.0 17.7 16.5 15.3 14.4 16.6 14.8 11.8 14.2 10.9 15.6 10.0 16.1 16.6 18.4 13.6 9.1 13.8 10.9 14.7 13.7 11.2 11.7 8.0 23.7 25.5 0.7 25.4
TDM (Ga) 2 2 2 2 2 2 2 2 2 2 1.2 1.0 2 2 2 1.8 2 2 2 1.2 2 2 2 2 2 2 2 2 2 2 2.8 2.9 2.3 2
*XRF analysis (data sources for Table 2); †LA-ICP-MS analysis (data sources for Table 2); ‡Analysis using isotope dilution method by TIMS. See text for method of 1Nd(550 Ma) and TDM calculations. Abbreviations as in Table 1.
Y. SUDA ET AL.
3 7 35 40 44 10 76 78 KY-22 KY-14 KY-15 KY-17 KY-19 KY-21 82 KY-11 87 89 86 KY-8 14 95 96 98 100 101 102 105 KY-5 KY-6 16 17 110 111 JB-1a
Location
¨ TZOW-HOLM COMPLEX METABASITES FROM LU
Fig. 4. Diagrams showing relation between modal abundance of biotite and whole-rock compositions of indicated elements and ratios for the mafic rocks of basaltic protoliths (Domain A in Fig. 5).
221
222
Y. SUDA ET AL.
Fig. 5. Mg-number v. SiO2/Al2O3 diagram for the metabasites. The grey field of ‘primitive’ basaltic magmas, the dashed arrow for differentiation trend of tholeiite, and the continuous arrows for the generalized direction of compositional change with accumulation of the indicated mineral phases are from Kempton & Harmon (1992) and Kempton et al. (1997).
whereas SiO2/Al2O3 ratio will decrease with accumulation of plagioclase. On this diagram, the mafic rocks are concentrated in a central domain (i.e. Domain A), around which the ultramafic rocks are distributed and plotted in four domains (i.e. Domains B, C, D and E). The domains almost correlate with the compositional trends related to the accumulation of Fe –Ti oxides, plagioclase and pyroxenes. Detailed fractional crystallization histories experienced by metabasites are discussed below. The C1 chondrite-normalized REE patterns and primitive mantle-normalized trace element patterns (spider diagrams) for metabasites from LHC are shown in Figure 6. The biotite-bearing samples are generally more enriched in REE and trace elements than the biotite-free samples. Exceptionally, the mafic rocks from Akebono Rock do not show a significant difference between the biotitefree and biotite-bearing samples, which suggests that whole-rock REE compositions might not have been affected by alkaline metasomatism. REE and trace element patterns for the mafic rocks from Akebono Rock are characterized by slight to moderate enrichment in light REE (LREE) (LaN/YbN ¼ 0.7–1.2; chondrite-normalized), and negative depletions for Ti and Nb (Fig. 6a). Those from Cape Hinode are characterized by moderate
enrichment in LREE (LaN/YbN ¼ 1.4–2.1), and negative anomalies for Ti, P and Nb (Fig. 6a). These features are comparable with those of island-arc basalt from Izu–Oshima (JB-2; Imai et al. 1995) and island-arc tholeiitic basalt in general (LaN/YbN ¼ 1.21 + 0.54, after Basaltic Volcanism Study Project 1981). Biotite-free mafic rocks from Skallevikshalsen, in the granulite-facies area, have profiles characterized by slight to moderate enrichment in REE (LaN/YbN ¼ 0.7– 1.8) and negative anomalies for P (Fig. 6c). These patterns are comparable with those of N-type (normal) to E-type MORB. The REE profiles and trace element patterns of biotite-bearing metabasites from the granulitefacies area are variable (Fig. 6b and c). Detailed discussions of these patterns are meaningless in considering the protolithic composition because of the possible effect of alkaline metasomatism, as described above. Profiles for the biotite-bearing metabasites from Rundva˚gshetta are characterized by significant negative anomalies of P and/or Ti, which may be a result of REE enrichment resulting from alkaline metasomatism. On the other hand, profiles for the biotite-bearing metabasites from Langhovde and Skallevikshalsen are similar to that of reference ocean island basalt (Fig. 6d). These metabasites have lower Mg-number and
¨ TZOW-HOLM COMPLEX METABASITES FROM LU
223
Fig. 6. Primitive mantle- and chondrite-normalized multi-element diagrams for the metabasites. (a) Mafic rocks from the amphibolite-facies area (Akebono Rock and Cape Hinode) and the transitional area (Akarui Point). (b) Mafic rocks from the granulite-facies area. (c) Ultramafic rocks from the granulite-facies area. (d) N-type MORB, ocean island basalt and E-type MORB from Sun & McDonough (1989), and island-arc basalt (JB-2) from Imai et al. (1995). Normalization values are from Sun & McDonough (1989). Data sources for REE are from LA-ICP-MS analysis, those for other elements are from XRF analysis (Table 2).
224
Y. SUDA ET AL.
SiO2/Al2O3 ratio than the primitive basaltic magma plotted in the field of Domain D (Fig. 5). Furthermore, they contain abundant Fe –Ti oxides and secondary biotite; hence they are considered as having a cumulate origin as a result of the fractionation of magnetite and/or ilmenite.
Sr and Nd isotopes Figure 7a shows 87Sr/86Sr ratios v. 143Nd/144Nd for metabasites from the LHC. The 87Sr/86Sr ratios increase with modal abundance of biotite, in that biotite-bearing samples have generally higher
Fig. 7. 87Sr/86Sr v. 143Nd/144Nd (a) and 143Nd/144Nd v. 147Sm/144Nd (b) diagrams for the metabasites. Slope of the grey band corresponds to an age of 1.5 Ga. Inset in (b) is an isochron diagram for mafic rocks from Akebono Rock. Isochron age and initial 143Nd/144Nd ratio (IR) were calculated using the computer program of Kawano (1994) with the equation of York (1966) and the decay constants 147Sm ¼ 6.54 10212 a21 from Lugmair & Marti (1978). The estimated relative errors in the age calculation of 147Sm/144Nd and 143Nd/144Nd ratios are 0.2% (1s) and 0.010% (1s), respectively.
ratios (0.70342–0.80029) than biotite-free samples (0.70313–0.73072). The whole-rock Rb concentrations in metabasites are strongly correlated with the modal abundance of secondary biotite, analogous to K2O (Fig. 4). Moreover, Rb, alkaline metal, and radiogenic 87Sr from 87Rb decay and 87Rb are rather mobile (e.g. Kagami et al. 2007). Therefore, the Sr isotopic ratios for metabasites must have been highly altered during the formation of secondary biotite or alkaline metasomatism. On the other hand, the 143Nd/144Nd ratios do not show a difference between the biotite-bearing samples (0.51126– 0.51314) and the biotite-free samples (0.51162– 0.51302). Thus, although the whole-rock REE contents of metabasites roughly increase with increasing modal abundance of biotite (Fig. 4), the 143Nd/144Nd ratios do not have significant correlation with the abundance of biotite. This result indicates that the Nd isotopic ratio was not altered during the formation of secondary biotite. The 87Sr/86Sr ratio for the metabasites increases with increasing metamorphic grade: 0.70313 – 0.71947 in the amphibolite-facies area; 0.70406 – 0.70474 in the transitional area; 0.70462–0.80029 in the granulite-facies area. The 143Nd/144Nd ratio also increases with increasing metamorphic grade: 0.51269 –0.51314 in the amphibolite-facies area; 0.51260– 0.51278 in the transitional area; 0.51126– 0.51284 in the granulite-facies area. Thus, both 87Sr/86Sr and 143Nd/144Nd ratios in metabasites are correlated with the metamorphic areas defined in the LHC, which could have two causes: (1) a variety of protolithic compositions for the metamorphic areas; (2) a degree of isotopic reequilibration by high-grade metamorphism of PanAfrican age. The Sm–Nd isochron diagram is shown in Figure 7b. Interpretation of these data is not straightforward because all the samples exhibit a great deal of scatter. However, the important point is that all the metabasites from the amphibolitefacies area and the transitional area are generally distributed around the 1.5 Ga reference line, indicating a Mesoproterozoic age. Moreover, although it is difficult to obtain identical isochron ages from the analysed samples because of the small range of 147Sm/144Nd ratios and the limited number of samples, an isochron age of 1497 + 214 Ma was tentatively obtained from the three samples from Akebono Rock (Fig. 7b; inset). Many geochronological studies have indicated that pre-Pan-African ages determined from the LHC are concentrated either at c. 1.5 Ga or 2.5 Ga. Therefore, the 1.5 Ga reference line in the isochron diagram possibly represents the protolithic age of the metabasites from the amphibolite-facies and transitional areas. On the other hand, the metabasites from the granulite-facies area show a great
¨ TZOW-HOLM COMPLEX METABASITES FROM LU
grate deal of scatter. In particular, the samples from Rundva˚gshetta have an extremely low 143Nd/144Nd ratio (0.51126–0.51156) compared with the others. Frost & Frost (1995) gave an example of re-equilibration of the whole-rock Sm –Nd isotopic system by open-system dehydration of amphibolite. Re-equilibration of Nd isotopes during metamorphism in rocks with homogeneous bulk Sm/Nd ratios produces a single horizontal array at the time of metamorphism, accompanied by an increase initial of Nd isotopic ratio. Thus, although an effect of isotopic re-equilibration during high-grade metamorphism cannot be totally ruled out, all the geochemical data, including 1Nd and model Nd ages, described below, do not show any evidence of isotopic re-equilibration of Pan-African age. The Nd isotopic compositions of the metabasites from the granulite-facies area must reflect variety of protolithic compositions. The Rb –Sr isotopic system is far more easily reset by metamorphism than the Sm–Nd isotopic system, and the closure temperature of the Rb –Sr whole-rock system generally decreases with higher activity of H2O (e.g. Kagami et al. 2003). Moreover, the possibility of subsurface exchange with seawater cannot be totally ruled out and could explain adequately the observed 87Sr enrichment in the metabasites. Therefore, the Sr isotopic compositions in metabasites probably reflect both a variety of protolithic compositions and multiple modifications by metamorphism and/or metasomatism.
Discussion Protoliths of metabasites On the Mg-number v. SiO2/Al2O3 diagram (Fig. 5), the majority of the mafic rocks plot in Domain A, which overlaps with the compositional field of ‘primitive’ basaltic magmas (Basaltic Volcanism Study Project 1981) and correlates with the tholeiitic fractionation trend. The Mg-number and SiO2/ Al2O3 ratio for the rocks of Domain A range from 34 to 65 and 2.8 to 4.1, respectively. The rocks are composed of relatively fine-grained minerals (,0.5 mm in length) with foliated or polygonal homogeneous textures (Fig. 3a). The main constituent minerals are hornblende, plagioclase, clinopyroxene and orthopyroxene. Accessory minerals are garnet, biotite, quartz, apatite and opaque minerals. The colour index ranges from 50 to 78, with an average of 64. These characteristics suggest that the mafic rocks in Domain A originated from basaltic rocks (i.e. basaltic protoliths). On the other hand, all the ultramafic rocks and some of the mafic rocks are distributed around the field of ‘primitive’ basaltic magma, and plot in
225
four definite domains (Fig. 5): Domains B, C, D and E. The rocks of Domain B have lower Mg-number values (34–56) and SiO2/Al2O3 ratios (1.7–3.3) than the field of ‘primitive’ basaltic magma, and correlate with a compositional trend referenced by Fe–Ti oxide accumulation. The rocks in this domain are composed of relatively coarse-grained minerals (,5 mm in length) with heterogeneous textures. The main constituent minerals are hornblende, plagioclase, clinopyroxene and orthopyroxene. Accessory minerals are garnet, biotite, quartz, apatite and opaque minerals, such as magnetite, ilmenite and pyrite. Opaque minerals (1–20% in modal abundance) and plagioclase (25–37% in modal abundance) are characteristically abundant (Fig. 3b). The opaque minerals occur both as inclusions in garnet and in the matrix. The colour index ranges from 43 to 90, with an average of 65. These features suggest that the metabasites of this domain originated from cumulates formed by a predominant accumulation of Fe–Ti oxides. The rocks of Domain C have a lower SiO2/ Al2O3 ratio (1.8–2.1), but almost the same Mg-number value (54– 64), compared with those of Domain A. This domain correlates with a compositional trend referenced by plagioclase accumulation. The rocks are composed of coarse-grained minerals (,5 mm in length) with heterogeneous textures. The main constituent minerals are hornblende, biotite, plagioclase and spinel. Accessory minerals are garnet and quartz. Plagioclase is characteristically abundant (34–80% in modal abundance). The colour index ranges from 20 to 66, with an average of 41. The rocks of this domain might have originated from cumulates formed by the predominant accumulation of plagioclase and spinel. The rocks of Domain D have higher Mg-number values (62–87) and lower SiO2/Al2O3 ratios (1.4– 3.7) than those of Domain A. The main constituent minerals are hornblende, clinopyroxene, orthopyroxene and plagioclase. Accessory minerals are garnet, biotite, spinel, sapphirine, corundum and opaque minerals. Either hornblende (20–90% in modal abundance) or pyroxene (0–70% in modal abundance) is abundant in the rocks. The colour index ranges from 57 to 97, with an average of 83. These properties suggest that the rocks have a cumulitic origin, formed by predominant accumulation of plagioclase and pyroxenes. The rocks of Domain E have higher Mg-number values (56– 88) and SiO2/Al2O3 ratios (4.1– 18) than those of Domain A, and correlate with a compositional trend referenced by pyroxene accumulation. The main constituent minerals are hornblende, clinopyroxene, orthopyroxene and plagioclase. Accessory minerals are olivine, biotite, spinel and opaque minerals (magnetite). The rocks
226
Y. SUDA ET AL.
in this domain are characterized by abundance of either hornblende (0– 72% in modal abundance) or pyroxenes (0–100% in modal abundance). Plagioclase is a minor phase, ranging in modal abundance from 0% to 42%, with an average of 17%. The colour index is the highest, ranging from 58 to 100, with an average of 83. All the metabasites containing olivine plot in this domain, and do not coexist with plagioclase. These characteristics suggest that the rocks of this domain originated from cumulates formed by predominant accumulation of olivine and pyroxenes. All these geochemical and petrographical characteristics, and the field evidence, suggest that the mafic rocks in Domain A had basaltic protoliths, whereas all the ultramafic rocks and some of the mafic rocks in other domains had cumulate protoliths. The regional contexts for the distribution of the rocks in Domains B, C, D and E (i.e. cumulitic protoliths) are not distinctive. They exclusively occur together in a narrow area. For example, rocks in both Domains D and E are occasionally found in a block of ultramafic rock at Akarui Point, forming a compositional layering of several metres thickness (Fig. 2d). This occurrence suggests a petrogenetic relation between the rocks. Moreover, the rocks in Domain A are distributed throughout the complex, and the majority of the mafic rocks from the amphibolite-facies area plot in Domain A, indicating a predominance of basaltic protoliths in this area.
Tectonic settings of basaltic protoliths Although Suda et al. (2006) have already indicated that mafic rocks from the amphibolite-facies area originated from basalts of tholeiitic affinities, whereas those from the transitional and granulitefacies areas originated from basalts of T-type to E-type MORB affinities, we would like to re-examine the tectonic setting for the metabasites of basaltic protolith on the basis that the basaltic protoliths must be limited to the rocks in Domain A, and some HFSE must have been mobilized by alkaline metasomatism after metamorphism. Figure 8 shows discrimination diagrams generally used to determine the tectonic setting of erupted basalts. In these diagrams, the biotitebearing samples exhibit a great deal of scatter, and have inconsistent geochemical characteristics between the diagrams. On the other hand, the biotite-free samples are consistent. The rocks from the amphibolite-facies area, Akebono Rock and Cape Hinode, plot in the field of island-arc basalt, whereas those from the granulite-facies area, Skallevikshalsen, mostly plot in the field of MORB. These results are also inferred from the profiles of primitive mantle-normalized trace
Fig. 8. Discrimination diagrams for the mafic rocks of basaltic protoliths (Domain A in Fig. 5). (a) Zr/Y v. Zr diagram after Pearce & Norry (1979). (b) V v. Ti/1000 diagram after Shervais (1982). (c) Cr v. Y diagram after Pearce (1982). Data are from XRF analysis (Table 2). IAB, island-arc basalt; MORB, mid-ocean ridge basalt; WPB, within-plate basalt.
¨ TZOW-HOLM COMPLEX METABASITES FROM LU
element patterns, as described above (Fig. 6b). Thus, the tectonic setting for the metabasites of basaltic protoliths differs between those in the amphibolite-facies area and in the granulitefacies area. The relative concentrations of Nb, Zr and Y of basalts, and the profiles of C1 chondritenormalized REE patterns are generally used to determine the type of MORB: N-type, T-type and E-type MORB. The profiles of C1-chondrite REE patterns indicate that mafic rocks of basaltic protoliths from the granulite-facies area, Skallevikshalsen, are of N-type to T-type MORB affinity (Fig. 6b). On the other hand, on the Nb–Zr– Y ternary diagram (Fig. 9), these plot in the field of N-type MORB, and are relatively enriched in Nb and Y compared with the typical compositions of N-type MORB (after Sun & McDonough 1989). Thus, the mafic rocks of basaltic protoliths from Skallevikshalsen are generally of N-type to T-type MORB affinity.
Magmatic processes of cumulitic protoliths Whole-rock transition metal concentrations provide strong evidence in evaluating the fractionation and accumulation processes of pyroxenes, olivine and Fe– Ti oxides from basaltic magma. For example, the Cr content is controlled by the presence of pyroxenes, and the Ni content by the presence of olivine. Chondrite-normalized transition metal patterns for metabasites from the transitional area are shown in Figure 10a. The rocks of Domains
Fig. 9. The Nb– Zr– Y discrimination diagram (Meschede 1986) for the mafic rocks of basaltic protoliths (Domain A in Fig. 5). Typical N-type MORB and E-type MORB compositions from Sun & McDonough (1989) are shown for comparison. Symbols as in Figure 8. Data are from XRF analysis (Table 2).
227
A and B in this area are characterized by relative depletions of Cr and Ni, compared with MORB from the mid-Atlantic-Ridge (Langmuir et al. 1977). On the other hand, the profiles for the rocks of Domains D and E are characterized by relative enrichment in Cr and Ni, and depletion in Ti and V showing a symmetrical relation with those for the rocks of Domains A and B. The Cr, Ni, Ti and V contents of all metabasites generally increase with increasing Mg-number values. Thus, these results suggest that the rocks in Domains D and E are the equivalent of cumulates, whereas the rocks in Domains A and B might have originated from their residual liquid. Therefore, the following fractionation processes are assumed in the metabasites from the transitional area. Olivine crystallized as the first phase with plagioclase, and formed cumulates corresponding to the rocks in Domain D. Subsequently, predominant accumulation of pyroxene formed cumulates corresponding to the rocks in Domain E. The rocks in Domain A are the equivalent of the residual liquid, from which Fe –Ti oxides crystallized as a later phase and formed cumulates corresponding to the rocks in Domain B. Many previous studies have indicated that olivine, spinel and calcic plagioclase are first to crystallize, and augite and Fe –Ti oxides are the later phases in typical ocean-floor basalt (e.g. Basaltic Volcanism Study Project 1981). On the other hand, calcic clinopyroxene becomes an early crystallizing phase under high-pressure and moderate PH2O conditions in some MORB and ocean island basalt (e.g. Bender et al. 1978). A fractional crystallization process for the metabasites from the transitional area as suggested above is consistent with the former case. Chondrite-normalized transition metal patterns for the metabasites from the granulite-facies area are shown in Figure 10b. In contrast to the rocks in the transitional area (Fig. 10a), the profiles for the rocks in Domains A and B in the granulitefacies area do not show a significant depletion in Cr and Ni. The CrN (chondrite-normalized) values are .0.0002. Only two samples from Domain D show significant depletion in Cr and Ni. Thus, although they occur in the same area or region, they do not show any evidence of a genetic relation among the rocks of cumulitic protoliths (i.e. Domains B, C, D, and E), and between those of basaltic protoliths (i.e. rocks from Domain A) and cumulitic protoliths. This remains to be investigated in a future study; however, these diagrams suggest that metabasites in the granulite-facies area have diverse origins and histories, and indicate that the geochemical characteristics of the metabasites clearly differ between the transitional area and the granulite-facies area; this will be further discussed in the next section.
228
Y. SUDA ET AL.
Fig. 10. Chondrite-normalized transition metal patterns for the metabasites from the transitional area (a) and granulitefacies area (b). Normalization values and grey fields for the MORB from Mid-Atlantic Ridge are from Langmuir et al. (1977). Data for Sc are from LA-ICP-MS analysis and those for other elements are from XRF analysis (Table 2).
Neodymium isotope constraints Model Nd (TDM) ages reflect the time elapsed since the sample was extracted from a given reservoir, usually
taken as the depleted mantle (DM). On the other hand, the epsilon notation, 1Nd(T), provides a comparison between the isotopic composition of sample and chondrite (Chondritic Uniform Reservoir; CHUR) at
¨ TZOW-HOLM COMPLEX METABASITES FROM LU
a given time. The TDM ages were calculated for the samples with 143Nd/144Nd ratio less than 0.5125 and 147Sm/144Nd ratio less than 0.13 using the DM parameters (¼0 Ga) and decay constant of 147Sm; 143 Nd/144Nd ¼ 0.513150, 147Sm/144Nd ¼ 0.2136, 147 l Sm ¼ 6.54 10212 a21, following the methods given by Kagami et al. (2006). The 1Nd(T) values were calculated following the equation given by DePaolo (1981): 1Nd(T) ¼ [143Nd/144Ndrock(T)/ 4 143 143Nd/144Nd Nd/144 NdCHUR CHUR(T) 2 1] 10 , (T) ¼ 0.511836 2 0.1967 [(exp lSmT) 2 1]. In this study, T was taken as 550 Ma, the time of most recent high-grade metamorphism in East
229
Gondwana, following the method of Harris et al. (1996). The results of these calculations are listed in Table 3. Figure 11 illustrates the distribution of the model Nd ages and the 1Nd(550 Ma) for metabasites from the LHC, in which the data for metamorphic rocks from the Rayner Complex (Black et al. 1987) and the Napier Complex (DePaolo et al. 1982; McCulloch & Black 1984; Black et al. 1986; Black & McCulloch 1987) are recalculated and shown for comparison. The Rayner Complex is considered as a part of a Mesoproterozoic mobile belt (e.g. Harley & Hensen 1990).
Fig. 11. Map showing distribution of Nd model (TDM) ages and 1Nd(550 Ma) for the metamorphic rocks from East Antarctica longitudes between 30 and 60 8E. Data from this study (*) and previous studies (**) are compiled in this map. Sources of data from previous studies are as follows: NC: DePaolo et al. (1982), McCulloch & Black (1984), Black et al. (1986) and Black & McCulloch (1987); RC: Black et al. (1987); LHC: Tanaka et al. (1985); Shiraishi et al. (1995); Nishi et al. (2002); Ajishi et al. (2004) and Kawano et al. (2005). NC, Napier Complex; RC, Rayner Complex; LHC, Lu¨tzow-Holm Complex; AF, amphibolite-facies area; TF, transitional area, N-GF, northern part of granulitefacies area, S-GF, southern part of granulite-facies area; OI, Ongul Islands; LA, Langhovde; SKL, Skallen; RVH, Rundva˚gshetta; CH, Cape Hinode; KR, Kasumi Rock; OR, Oku-iwa Rock; SVN, Skarvsnes.
230
Y. SUDA ET AL.
Orthogneisses from this complex are characterized by negative 1Nd(550 Ma) values (219 to 23.5) with Archaean TDM ages (3.6 –2.5 Ga). The Napier Complex is considered as a part of an Early Archaean craton (e.g. Harley & Hensen 1990). Orthogneisses and mafic granulites from this complex are characterized by pronounced negative 1Nd(550 Ma) values (250 to 219) with the oldest TDM ages (.4.1 Ga). Results of the model Nd ages for metabasites from the LHC indicate that relatively young model ages between 1.0 and 1.8 Ga are distributed around the northern part of the granulite-facies area of the LHC (i.e. Ongul Islands, Langhovde and Skallen areas), whereas older ages between 2.3 and 2.9 Ga are distributed in the southern part of the granulite-facies area (i.e. Rundva˚gshetta area). The younger ages are almost equivalent to the model Nd ages estimated from the quartzofeldspathic gneisses and granitic gneisses (1.0–1.4 Ga; recalculated based on the data given in earlier studies: Skarvsnes, Tanaka et al. 1985; Cape Hinode, Shiraishi et al. 1995; Oku-iwa Rock, Nishi et al. 2002; Kasumi Rock, Ajishi et al. 2004; Ongul Islands, Kawano et al. 2005). On the other hand, the 1Nd(550 Ma) values for metabasites decrease systematically to the SW across the complex: þ18 to þ23 in the amphibolite-facies area; þ17 to þ20 in the transitional area; þ13 to þ18 in the northern part of the granulite-facies area; 25.5 to þ0.7 in the southern part of the granulite-facies area. This trend is also seen in the quartzo-feldspathic and granitic gneisses from the LHC. Thus, the isotope data from the LHC indicate two age provinces: a younger amphibolite–granulite terrane (TDM ¼ 1.0–1.8 Ga and 1Nd(550 Ma) ¼ þ8.0 to þ23), and an older granulite terrane (TDM ¼ 2.3–2.9 Ga and 1Nd(550 Ma) ¼ 25.5 to þ0.7). Moreover, all these results indicate that the LHC is composed of relatively young crustal materials derived from active crustal growth during Mesoproterozoic time, and older crustal materials that may be derived from Palaeoproterozoic to Archaean basement of craton and oceanic crust. The LHC is formed from an assembly of tectonic blocks or terranes of different origins. The blocks must have been amalgamated by multiple subduction of Pan-African and/or pre-Pan-African age, and they underwent high-grade metamorphism during the final stages of collision of East and West Gondwana. This occurrence must represent the process of continental growth of Gondwana. Recent work by Kemp et al. (2006) suggested that crustal generation of the Gondwana supercontinent is limited to major pulses at 1.9 and 3.3 Ga. Furthermore, global peaks of igneous activity derived from mantle superplumes and rapid crustal generation are concentrated at 2.7, 1.9 and 1.2 Ga (Stein & Hofmann
1994; Albare`de 1998; Condie 1998). The results of Nd model ages in this study also suggest that the metabasites in the LHC have been generated during these global peaks of igneous activity and crustal formation.
Conclusions Biotite-bearing metabasites in the Lu¨tzow-Holm Complex had experienced alkaline metasomatism during cooling. Consequently, whole-rock concentrations of alkaline elements and some HFSE (Nb, Zr, Ti and REE) had been altered and seldom represent protolithic signatures. The Sr isotopes had also been altered by the introduction of the Rb. On the other hand, whole-rock Mg-number and SiO2/Al2O3 ratio have not been affected by the alkaline metasomatism, which may represent protolithic conditions. The whole-rock Mg-number and SiO2/Al2O3 ratio suggest that the majority of mafic rocks have basaltic protolith compositions with tholeiitic affinity, whereas all the ultramafic rocks and some of the mafic rocks have a cumulitic protolith. This result is also inferred from the field occurrences and petrography. The basaltic protoliths generally occur as continuous layers or lenses, which are composed of fine-grained minerals with a homogeneous texture. On the other hand, the cumulitic protoliths generally occur as blocks with several kinds of compositional layering, which are composed of relatively coarse-grained minerals with a heterogeneous texture. Geochemical characteristics of the biotite-free metabasites indicate that the basaltic protoliths from the amphibolite-facies area have an arc affinity, whereas those from the granulite-facies area have N-type and T-type MORB (i.e. ocean-floor) affinities. The cumulitic protoliths from the transitional area had a fractional crystallization history as follows: olivine þ plagioclase accumulations and subsequent pyroxene and Fe–Ti oxides accumulation. The Nd isotopic data suggest that the LHC is composed of a younger amphibolite –granulite terrane and an older granulite terrane. Crustal materials in the younger terrane may be derived from active crustal growth during Mesoproterozoic time, whereas those in the older terrane may be derived from a Palaeoproterozoic to Archaean craton and oceanic crust. That is, the LHC is formed from an assembly of tectonic blocks of diverse origins and ages. The blocks must have been amalgamated by multiple subduction of Pan-African and/or pre-Pan-African age, and they underwent high-grade metamorphism during the final collision of East and West Gondwana.
¨ TZOW-HOLM COMPLEX METABASITES FROM LU We are very grateful to T. Morishita and to A. I. S. Kemp for constructive reviews of an earlier version of the manuscript. We would like to express our sincere thanks to H. Kagami and M. Sato for mass spectrometry analysis at Niigata University. Y. Motoyoshi, M. Satish-Kumar, S. Kagashima, N. Ishikawa, M. Arita, K. Naito, T. Ikeda, T. Kawakami and T. Kawasaki are thanked for collaboration in fieldwork. Thanks are also given to Y. Osanai, H. Ishizuka, M. Owada, T. Nakajima, S. Miyashita, Y. Hayasaka, T. Miyamoto, K. Shiraishi, T. Hokada, D. J. Dunkley and Y. Nogi for helpful guidance, comments and discussions.
References A JISHI , H., K AWANO , Y., K AWAKAMI , T. & I KEDA , T. 2004. Geochronological study of post-metamorphic granite from Kasumi Rock, Lu¨tzow-Holm Complex, East Antarctica. Polar Geoscience, 17, 35–44. A LBARE` DE , F. 1998. The growth of continental crust. Tectonophysics, 296, 1 –14. Basaltic Volcanism Study Project 1981. Basaltic Volcanism on the Terrestrial Planets. Pergamon, New York. B ENDER , J. F., H ODGES , F. N. & B ENCE , A. E. 1978. Petrogenesis of basalts from the project FAMOUS area: experimental study from 0 to 15 kbars. Earth and Planetary Science Letters, 41, 277 –302. B LACK , L. P. & M C C ULLOCH , M. T. 1987. Evidence for isotopic equilibration of Sm– Nd whole-rock systems in early Archean crust of Enderby Land, Antarctica. Earth and Planetary Science Letters, 82, 15– 24. B LACK , L. P., S HERATON , J. W. & J AMES , P. R. 1986. Late Archean granites of the Napier Complex, Enderby Land, Antarctica: a comparison of Rb– Sr, Sm– Nd and U –Pb isotopic systematics in a complex terrain. Precambrian Research, 32, 343–368. B LACK , L. P., H ARLEY , S. L., S UN , S. S. & M C C ULLOCH , M. T. 1987. The Rayner Complex of East Antarctica: complex isotopic systematics within a Proterozoic mobile belt. Journal of Metamorphic Geology, 5, 1–26. B UCHER , K. & F REY , M. 1994. Petrogenesis of Metamorphic Rocks, 6th edn. Springer, New York. C ONDIE , K. C. 1998. Episodic continental growth and supercontinents: A mantle avalanche connection? Earth and Planetary Science Letters, 163, 97–108. D E P AOLO , D. J. 1981. Neodymium isotopes in the Colorado Front Range and crust –mantle evolution in the Proterozoic. Nature, 291, 193–196. D E P AOLO , D. J., M ANTON , E. S., G REW , E. S. & H ALPERN , M. 1982. Sm– Nd, Rb– Sr and U–Th– Pb systematics of granulite facies rocks from Fyfe Hills, Enderby Land, Antarctica. Nature, 298, 614–619. E GGINS , S. M. 2003. Laser ablation ICP-MS analysis of geological materials prepared as lithium borate glasses. Geostandards Newsletter, 27, 147–162. F ITZSIMONS , I. C. W. 1996. Metapelitic migmatites from Brattstrand Bluffs, East Antarctica—metamorphism, melting and exhumation of the mid crust. Journal of Petrology, 37, 395– 414. F ITZSIMONS , I. C. W. 2000. A review of tectonic events in the East Antarctic Shield and their implications for
231
Gondwana and earlier supercontinent. Journal of African Earth Sciences, 31, 3 –23. F LOYD , P. A. & W INCHESTER , J. A. 1978. Identification and discrimination of altered and metamorphosed volcanic rocks using immobile elements. Chemical Geology, 21, 291– 306. F ROST , C. D. & F ROST , B. R. 1995. Open-system dehydration of amphibolite, Morton Pass, Wyoming: elemental and Nd and Sr isotopic effects. Journal of Geology, 103, 269–284. G RUNOW , A., H ANSON , R. & W ILSON , T. 1996. Were aspects of Pan-African deformation linked to lapetus opening? Geology, 24, 1063– 1066. H ARLEY , S. L. & H ENSEN , B. J. 1990. Archean and Proterozoic high-grade terranes of East Antarctica (40– 808E): a case study of diversity in granulite facies metamorphism. In: A SHWORTH , J. R. & B ROWN , M. (eds) High-temperature Metamorphism and Crustal Anatexis. Unwin Hyman, London, 320–370. H ARRIS , N. B. W., B ARTLETT , J. M. & S ANTOSH , M. 1996. Neodymium isotope constraints on the tectonic evolution of East Gondwana. Journal of Southeast Asian Earth Sciences, 14, 119–125. H IROI , Y., S HIRAISHI , K., N AKAI , Y., K ANO , T. & Y OSHIKURA , S. 1983. Geology and petrology of Prince Olav Coast, East Antarctica. In: O LIVER , R. L., J AMES , P. R. & J AGO , J. B. (eds) Antarctic Earth Science. Australian Academy of Science, Canberra, A.C.T., 32–35. H IROI , Y., S HIRAISHI , K., M OTOYOSHI , Y., K ANISAWA , S., Y ANAI , K. & K IZAKI , K. 1986. Mode of occurrence, bulk chemical compositions, and mineral textures of ultramafic rocks in the Lu¨tzow-Holm Complex, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 43, 62–84. H IROI , Y., S HIRAISHI , K. & M OTOYOSHI , Y. 1991. Late Proterozoic paired metamorphic complexes in East Antarctica, with special reference to the tectonic significance of ultramafic rocks. In: T HOMSON , M. R. A., C RAME , J. A. & T HOMSON , J. W. (eds) Geological Evolution of Antarctica. Cambridge University Press, Cambridge, 83– 87. H IROI , Y., M OTOYOSHI , Y., S ATISH -K UMAR , M., K AGASHIMA , S., S UDA , Y. & I SHIKAWA , N. 2006. Granulites from Cape Hinode in the amphibolitefacies eastern part of Prince Olav Coast, East Antarctica: new evidence for allochthonous block in the Lu¨tzow-Holm Complex. Polar Geoscience, 19, 89–108. H YNES , A. 1980. Carbonatization and mobility of Ti, Y, and Zr in Ascot Formation metabasalts, SE Quebec. Contributions to Mineralogy and Petrology, 75, 79–87. I MAI , N., T ERASHIMA , S., I TOH , S. & A NDO , A. 1995. 1994 compilation of analytical data for minor and trace elements in seventeen GSJ geochemical reference samples, ‘Igneous Rock Series’. Geostandards Newsletter, 19, 135– 213. J ACKSON , J. A. 1997. Glossary of Geology, 4th edn. American Geological Institute, Alexandria, VA. J ACOBS , J. & T HOMAS , R. J. 2004. Himalayan-type indenter-escape tectonics model for the southern part of the late Neoproterozoic– early Paleozoic East African–Antarctic orogen. Geology, 32, 721–724.
232
Y. SUDA ET AL.
J OHANNES , W. & H OLTZ , F. 1996. Petrogenesis and Experimental Petrology of Granitic Rocks. Springer, Berlin. K AGAMI , H., S HIMURA , T., Y UHARA , M., O WADA , M., O SANAI , Y. & S HIRAISHI , K. 2003. Resetting and closing condition of Rb–Sr whole-rock isochron system: some samples of metamorphic and granitic rocks from the Gondwana super-continent and Japan Arc. Polar Geoscience, 16, 227– 242. K AGAMI , H., K AWANO , Y., A KIYAMA , M. ET AL . 2006. Provenance of Paleozoic–Mesozoic sedimentary rocks in the Inner Zone of Southwest Japan: An evaluation based on Nd model ages. Gondwana Research, 9, 142– 151. K AGAMI , H., I MAOKA , T., I SHIOKA , J. ET AL . 2007. Issues relating to Rb– Sr and Sm–Nd whole rock isochrons. MAGMA, 87, 1–29 [in Japanese with English abstract]. K AKUBUCHI , S., I SHISAKA , T. & K AWANO , Y. 1999. X-ray fluorescence analysis of major and trace elements in silicate rocks using fused disk samples. Journal of the Faculty of Culture and Education, Saga University, 3, 133– 140 [in Japanese with English abstract]. K ANISAWA , S., Y OSHIDA , T. & I SHIKAWA , K. 1987. Preliminary study on trace element geochemistry of metabasites around Syowa Station and the adjacent areas. Proceedings of NIPR Symposium on Antarctic Geosciences, 1, 98– 106. K AWANO , Y. 1994. Calculation program for isochron ages of Rb –Sr and Sm– Nd systems using personal computer. Geoinformatics, 5, 13–19 [in Japanese with English abstract]. K AWANO , Y., N ISHI , N. & I SHISAKA , T. 1999. Preparation of samples for Sr and Nd isotopic analysis at petrochemical laboratory. Journal of the Faculty of Culture and Education, Saga University, 4, 139–146 [in Japanese with English abstract]. K AWANO , Y., M ENO , A., N ISHI , N. & K AGAMI , H. 2005. Geochemistry of the pre/syn-metamorphic granite in the Ongul Islands, East Antarctica. Polar Geoscience, 18, 114 –129. K EMP , A. I.S., H AWKESWORTH , C. J., P ATERSON , B. A. & K INNY , P. D. 2006. Episodic growth of the Gondwana supercontinent from hafnium and oxygen isotopes in zircon. Nature, 439, 580– 583. K EMPTON , P. D. & H ARMON , R. S. 1992. Oxygen isotope evidence for large-scale hybridization of the lower crust during magmatic underplating. Geochimica et Cosmochimica Acta, 56, 971– 986. K EMPTON , P. D., D OWNES , H. & E MBEY -I SZTIN , A. 1997. Mafic granulite xenoliths in Neogene alkali basalts from the western Pannonian Basin: insights into the lower crust of a collapsed orogen. Journal of Petrology, 38, 941–970. K RIEGSMAN , L. M. 1995. The Pan-African event in East Antarctica: a view from Sri Lanka and the Mozambique Belt. Precambrian Research, 75, 263–277. L ANGMUIR , C. H., B ENDER , J. F., B ENCE , A. E., H ANSON , G. N. & T AYLOR , S. R. 1977. Petrogenesis of basalts from the FAMOUS area: Mid-Atlantic Ridge. Earth and Planetary Science Letters, 36, 133–156. L UGMAIR , G. W. & M ARTI , K. 1978. Lunar initial 143 Nd/144Nd: differential evolution of the lunar crust
and mantle. Earth and Planetary Science Letters, 39, 349–357. M C C ULLOCH , M. T. & B LACK , L. P. 1984. Sm–Nd isotopic systematics of Enderby Land granulites and evidence for the redistribution of Sm and Nd during metamorphism. Earth and Planetary Science Letters, 71, 46–58. M EERT , J. G. 2003. A synopsis of events related to the assembly of eastern Gondwana. Tectonophysics, 362, 1– 40. M ELFI , A. J., F IGUEIREDO , A. M., K RONBERG , B. I., D OHERT , W. D. & M ARQUES , L. S. 1990. REE mobilities during incipient weathering of volcanic rocks of the Parana Basin, Brazil. Chemical Geology, 84, 375–376. M ESCHEDE , M. 1986. A method of discriminating between different types of mid-ocean-ridge basalts and continental tholeiites with the Nb– Zr– Y diagram. Chemical Geology, 56, 207–218. M IDDLEMOST , E. A. K. 1989. Iron oxidation ratios, norms and the classification of volcanic rocks. Chemical Geology, 77, 19–26. M OTOYOSHI , Y. & I SHIKAWA , M. 1997. Metamorphic and structural evolution of granulites from Rundva˚gshetta, Lu¨tzow-Holm Bay, East Antarctica. In: R ICCI , C. A. (ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antartica, Siena, 65–72. M OTOYOSHI , Y. & S HIRAISHI , K. 1994. Quantitative chemical analyses of rocks with X-ray fluorescence analyzer: (1) Major elements. Antarctic Record, 39, 40–48 [in Japanese with English abstract]. M OTOYOSHI , Y., I SHIZUKA , H. & S HIRAISHI , K. 1996. Quantitative chemical analyses of rocks with X-ray fluorescence analyzer: (2) Trace elements. Antarctic Record, 40, 53–63 [in Japanese with English abstract]. N ISHI , N., K AWANO , Y. & K AGAMI , H. 2002. Rb– Sr and Sm– Nd isotopic geochronology of the granitoid and hornblende biotite gneiss from the Oku-iwa Rock in the Lu¨tzow-Holm Complex, East Antarctica. Polar Geoscience, 15, 46–65. P EARCE , J. A. 1982. Trace element characteristics of lavas from destructive plate boundaries. In: T HORPE , R. S. (ed.) Andesites. Wiley, Chichester, 525– 548. P EARCE , J. A. 2002. Supra-subduction zone ophiolites: The search for modern analogs. In: D ILEK , Y. & N EWCOMB , S. (eds) Ophiolite Concept and the Evolution of Geological Though. Geological Society of America, Special Papers, 373, 269– 229. P EARCE , J. A. & N ORRY , M. J. 1979. Petrogenetic implications of Ti, Zr, Y, and Nb variations in volcanic rocks. Contributions to Mineralogy and Petrology, 69, 33–47. S ATISH -K UMAR , M., K AGASHIMA , S., S UDA , Y. & M OTOYOSHI , Y. 2006a. Geology of Byoˆbu Rock and Gobanme Rock, Prince Olav Coast, East Antarctica. Polar Geoscience, 19, 1– 36. S ATISH -K UMAR , M., M OTOYOSHI , Y., S UDA , Y., H IROI , Y. & K AGASHIMA , S. 2006b. Calc-silicate rocks and marbles from Lu¨tzow-Holm Complex, East Antarctica, with special reference to the mineralogy and geochemical characteristics of calc-silicate meta-boudins from Rundva˚gshetta. Polar Geoscience, 19, 37–61.
¨ TZOW-HOLM COMPLEX METABASITES FROM LU S HERVAIS , J. W. 1982. Ti–V plots and the petrogenesis of modern and ophiolitic lavas. Earth and Planetary Science Letters, 59, 101– 118. S HIRAISHI , K., E LLIS , D. J., H IROI , Y., F ANNING , Y., M OTOYOSHI , Y. & N AKAI , Y. 1994. Cambrian orogenic belt in East Antarctica and Sri Lanka: implications for Gondwana assembly. Journal of Geology, 102, 47–65. S HIRAISHI , K., K AGAMI , H. & Y ANAI , K. 1995. Sm–Nd and Rb –Sr isochron ages for meta-trondhjemites from Cape Hinode, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 8, 130–136. S HIRAISHI , K., H OKADA , T., F ANNING , C. M., M ISAWA , K. & M OTOYOSHI , Y. 2003. Timing of thermal events in eastern Dronning Maud Land, East Antarctica. Polar Geoscience, 16, 76–99. S TEIN , M. & H OFMANN , A. W. 1994. Mantle plumes and episodic crustal growth. Nature, 372, 63– 68. S UDA , Y., K AGASHIMA , S., S ATISH -K UMAR , M., M OTOYOSHI , Y. & H IROI , Y. 2006. Geochemistry of mafic metamorphic rocks in the Lu¨tzow-Holm Complex, East Antarctica: implications for tectonic evolution. Polar Geoscience, 19, 62–88. S UN , S.-S. & M C D ONOUGH , W. F. 1989. Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In: S AUNDERS , A. D. & N ORRY , M. J. (eds) Magmatism in the Ocean Basins. Geological Society, London, Special Publications, 42, 313– 345.
233
T ANAKA , T., N AKAJIMA , T. & S HIRAISHI , K. 1985. Sm–Nd ages on the metamorphic rocks from Skarvsnes, East Antarctica. 6th Symposium on Antarctic Geoscience Abstracts, 20 [in Japanese]. T ANAKA , T., T OGASHI , S., K AMIOKA , H. ET AL . 2000. JNdi-1: a neodymium isotopic reference in consistency with La Jolla neodymium. Chemical Geology, 168, 279– 281. U LIANOV , A., K ALT , A. & P ETTKE , T. 2006. Aluminous websterite and granulite xenoliths from the Chyulu Hills volcanic field, Kenya: gabbro–troctolitic cumulates subjected to lithospheric foundering. Contributions to Mineralogy and Petrology, 152, 459–483. V ALSAMI , E. & C ANN , J. R. 1992. Mobility of rare earth element in zones of intense hydrothermal alteration in the Pindos ophiolite, Greece. In: P ARSON , L. M., M URTON , B. J. & B ROWNING , P. (eds) Ophiolites and their Modern Oceanic Analogues. Geological Society, London, Special Publications, 60, 219– 232. W ILSON , T. J., G RUNOW , A. M. & H ANSON , R. E. 1997. Gondwana assembly: the view from Southern Africa and East Gondwana. Journal of Geodynamics, 23, 263– 286. Y ORK , D. 1966. Least-squares fitting of a straight line with correlated errors. Canadian Journal of Physics, 44, 1079–1084. Y OSHIDA , M. 1978. Tectonics and petrology of charnockites around Lu¨tzow-Holmbukta, East Antarctica. Journal of Geosciences, Osaka City University, 21, 65–152.
Geochemistry of post-kinematic mafic dykes from central to eastern Dronning Maud Land, East Antarctica: evidence for a Pan-African suture in Dronning Maud Land MASAAKI OWADA1, SOTARO BABA2, YASUHITO OSANAI3 & HIROO KAGAMI4 1
Department of Earth Sciences, Yamaguchi University, 1677-1 Yoshida, Yamaguchi, 753-8512 Japan (e-mail:
[email protected]) 2
Department of Natural Environment, University of the Ryukyus, 1 Senbaru, Nishihara, Okinawa, 903-0213 Japan
3
Division of Evolution of Earth Environments, Graduate School of Social and Cultural Studies, Kyushu University Ropponmatsu 4-2-1, Fukuoka, 810-8560 Japan 4
Graduate School of Science and Technology, Niigata University, Niigata, 950-2181 Japan Abstract: The region comprising central to eastern Dronning Maud Land (28W to 408E), East Antarctica, is underlain by Mesoproterozoic to Cambrian metamorphic rocks and post-kinematic intrusive rocks with varied compositions. The post-kinematic mafic dykes linked to the Pan-African orogen include various types of lithologies: lamprophyre and lamproite in Mu¨hligHofmannfjella in central Dronning Maud Land and lamprophyre and high-K dolerite in the Sør Rondane Mountains in eastern Dronning Maud Land. Most of the mafic dykes have been weakly affected by low-grade metamorphism, but clearly preserve their igneous textures. The mafic dykes show a high abundance of Rb, Ba, Sr and light rare earth elements with negative anomalies of Nb, Ta and Ti in a multi-element primitive mantle-normalized diagram. The geochemical characteristics of the mafic dykes suggest that they were derived from a metasomatized mantle source leaving phlogopite, rutile and/or titanite as residual phases. Considering Sr and Nd isotopic systematics of the mafic dykes and the host metamorphic rocks and coeval felsic intrusive rocks, a large crustal boundary potentially related to a suture zone of West and East Gondwana should pass between Mu¨hlig-Hofmannfjella and the Sør Rondane Mountains.
High-K mafic magmas including lamprophyre and lamproite are generally thought to be derived from the lithospheric mantle; these have a relatively small degree of melting and appear in various tectonic settings (Rock 1987; Foley et al. 1987). The mantle source involved in the generation of high-K magmas requires at least two different components—a peridotitic reservoir and a component enriched in large ion lithophile elements (LILE) and light rare earth elements (LREE)—because the primitive magmas of these rocks are characterized by the high abundance of both compatible and incompatible elements. The mantle source for high-K magmas is probably produced by metasomatic processes in the lithosphere mantle regardless of its formation age (Peterson et al. 2002). Therefore, the contents of incompatible elements and the isotopic signatures of high-K mafic rocks should reflect the compositions of the lithospheric sources. The high-K mafic dykes in the region comprising central to eastern Dronning Maud Land (28W to 408E), East Antarctica (Fig. 1), are related to the post-orogenic igneous activity after
the Pan-African continental collision. They intruded the Mesoproterozoic to Cambrian metamorphic terranes. Ikeda et al. (1995) reported the geochemistry of the major and trace elements and the Sr and Nd isotopic compositions of the high-K mafic dykes from the Sør Rondane Mountains, eastern Dronning Maud Land. They concluded that these dykes were produced by the partial melting of the metasomatized mantle sources. All the high-K mafic dykes in the Sør Rondane Mountains are believed to have originated from similar mantle sources based on their geochemical features, which include the Sr and Nd isotope systematics. Hoch et al. (2001) described the geochemical features involving the Sr, Nd, Pb and O isotopes of high-K dykes from the Schirmacher Oasis (708440 S – 708470 S, 118250 E–118550 E; Fig. 1) in central Dronning Maud Land and discussed the petrogenesis of minettes, which were derived from the metasomatized lithosphere. However, the Sr and Nd isotopes of the mafic dykes from the Sør Rondane Mountains are completely different from those of the minettes in the Schirmacher Oasis; thus the Sør Rondane
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 235 –252. DOI: 10.1144/SP308.12 0305-8719/08/$15.00 # The Geological Society of London 2008.
236
M. OWADA ET AL.
Fig. 1. Location of central to eastern Dronning Maud Land.
Mountains and Schirmacher Oasis could belong to a segmented post-collisional high-K magmatic belt. In the plate-tectonic reconstructions of the Gondwana supercontinent, the Dronning Maud Land mountain range is inferred to represent the southeastern continuation of the East African Orogen (Stern 1994; Jacobs et al. 1998, 2003; Bauer et al. 2003). The East African –East Antarctic Orogen was formed as a result of the collision between East and West Gondwana during the Pan-African event in late Neoproterozoic to early Palaeozoic time. Before the collision, it is inferred that the Mozambique Ocean extended into the Lu¨tzow-Holm Bay area (Fig. 1) (Shiraishi et al. 1994). Although the Mozambique suture resulted in the closure of this ocean, the position of the suture, which is inferred to pass between 28W and 148E (Bauer et al. 2003) or 258E and 458E (Fitzsimons 2000; Boger et al. 2001; Asami et al. 2005), is still controversial. The present study expands the geochemical study of Ikeda et al. (1995) with additional samples and highlights the geochemical features of the mafic dykes in the region extending from central to eastern Dronning Maud Land. A newly found lamproite dyke in central Dronning Maud Land (718550 S, 88220 E) is also used for this geochemical study. Finally, this paper discusses the tectonic implications of the position of the segmented boundary that potentially corresponds to the suture of West and East Gondwana.
General geology of central to eastern Dronning Maud Land The coastal mountain range in the region extending from central to eastern Dronning Maud Land is exposed subparallel to the East Antarctica Ice Sheet and is c. 200 –250 km inland of the edge of the ice sheet (Fig. 1). The older Mesoproterozoic
(c. 1.1Ga) basement rocks in Dronning Maud Land were differentially reworked during the Pan-African event. The Pan-African tectonothermal overprint was less intense in western Dronning Maud Land but strongly affected the Mesoproterozoic rocks in the region extending from central to eastern Dronning Maud Land. Moreover, the eastern end of Dronning Maud Land, the Lu¨tzow-Holm Complex, comprises juvenile Neoproterozoic crusts, and it was affected only by the Pan-African orogeny (Shiraishi et al. 1994). Voluminous granitoids are exposed over a large area, extending from 28E in the west to 288E in the east. These c. 500 Ma granitoids are mainly undeformed, except in a few localized shear zones. The lithology of the postkinematic intrusive rocks and a summary of their age data are listed in Table 1. The mountains and nunataks of central Dronning Maud Land (2–148E; hereafter referred to as the Mu¨hlig-Hofmann region; Fig. 1) consist of a series of post-tectonic intrusive rocks, which are emplaced into high-grade metamorphic basement rocks (Dallmann et al. 1990; Austrheim et al. 1997; Ohta 1999; Jacobs & Bauer 2001; Owada et al. 2003) (Fig. 2a). These basement rocks comprise banded gneisses and migmatites of various compositions, whereas the igneous suite includes voluminous intrusions of charnockite, syenite, quartz-syenite, granite and several generations of dykes. The compositions of dyke intrusions vary from granitic compositions to gabbroic or dioritic compositions and include lamprophyre and lamproite. Some of the granitoids are characterized by Fe-enriched bulk composition and contain fayalite (Ohta et al. 1990). Age dating in the Mu¨hlig-Hofmann region reveals the presence of two tectonothermal events. Cooling and intrusive ages around 400–500 Ma were reported by Ravich & Krylov (1964) and Ohta et al. (1990) by using the K– Ar and Rb– Sr isotopic data. The Rb–Sr and Sm–Nd isotope data
GEOCHEMISTRY OF POST-KINEMATIC MAFIC DYKE
237
Table 1. Summary of age determinations for igneous rocks from central to eastern Dronning Maud Land Lithology, occurrence Mu¨hlig-Hofmann region Micro-granite, dyke Aplite, dyke Charnockite, batholith Syenite, batholith Syenite, batholith Lamprophyre, dyke Granodiorite, stock Anorthosite, stock Anorthosite, stock Charnockite, stock Metagranite, basement Metagranite, basement Sør Rondane Mountains Dolerite, dyke Dolerite, dyke Dolerite, dyke Dolerite, dyke Dolerite, dyke Granite, stock Granite, stock Granite, stock Granite, stock Meta-tonalite, basement Enderbitic gneiss, basement Enderbitic gneiss, basement
Method
Age (Ma)
Reference
SHRIMP, zircon TIMS, zircon Rb –Sr, whole-rock isochron SIMS, zircon TIMS, zircon SHRIMP, zircon SHRIMP, zircon SHRIMP, zircon SHRIMP, zircon SHRIMP, zircon SHRIMP, zircon SHRIMP, zircon
486 + 4 495 + 7 500 + 24 500 + 8* 501 + 10* 523 + 5 530 + 8 583 + 7 600 + 12 608 + 9 1086 + 20 1087 + 28
Jacobs et al. (2003) Paulsson & Austrheim (2003) Ohta et al. (1990) Paulsson & Austrheim (2003) Paulsson & Austrheim (2003) Jacobs et al. (2003) Jacobs et al. (1998) Jacobs et al. (1998) Jacobs et al. (1998) Jacobs et al. (1998) Jacobs et al. (1998) Jacobs et al. (1998)
Ar –Ar, whole-rock K –Ar, whole-rock Ar –Ar, whole-rock K –Ar, whole-rock K –Ar, whole-rock Rb –Sr, whole-rock isochron Rb –Sr, whole-rock isochron Sm –Nd, whole-rock isochron Rb –Sr, whole-rock isochron Rb – Sr, whole-rock isochron Sm –Nd, whole-rock isochron Rb –Sr, whole-rock isochron
476 + 23 467 + 7 439 + 13* 451 + 12* 488 + 18* 506 + 43 528 + 31* 519 + 98* 525 + 32 956 + 39 961 + 101* 978 + 52*
Takigami et al. (1987) Takigami et al. (1987) Takigami et al. (1987) Takigami et al. (1987) Takigami et al. (1987) Tainosho et al. (1992) Tainosho et al. (1992) Arakawa et al. (1994) Takahashi et al. (1990) Takahashi et al. (1990) Shiraishi & Kagami (1992) Shiraishi & Kagami (1992)
*The same sample determined by different methods. TIMS, thermal ionization mass spectrometry. TIMS, thermal ionization mass spectrometry.
presented by Moyes (1993) indicated an intrusive age of 1153 Ma for the granitic gneisses in Gjelsvikfjella. U–Pb dating of zircon also detected two events (Jacobs et al. 1998, 2003; Paulsson & Austrheim 2003), which occurred before c. 1100 Ma and between 560 and 490 Ma. The zircon cores from granitic migmatite in Gjelsvikfjella yielded a protolith age of 1163 + 6 Ma, whereas metamorphic zircon rims recorded a recrystallization age of 504 + 6 Ma (Paulsson & Austrheim 2003). Jacobs et al. (2003) divided the latter event into two: the Pan-African I event (c. 560 Ma) and the PanAfrican II event (c. 530– 490 Ma). The Pan-African I event corresponds to the collision of West and East Gondwana and is associated with intense deformation, whereas the Pan-African II event is characterized by little deformation and the emplacement of voluminous intrusive rocks over a period of c. 30 Ma between 520 and 490 Ma (Jacobs et al. 2003). An undeformed lamprophyre dyke with a zircon crystallization age of 523 + 5 Ma indicates the absence of deformation or the presence of minor deformation after its intrusion (Jacobs et al. 2003). Post-kinematic syenite and aplite in the same area were dated to
501 + 10 Ma and 495 + 14 Ma, respectively (Paulsson & Austrheim 2003). Minette dykes from the Schirmacher Oasis located in the coastal region (Fig. 1) intrude the Precambrian basement rocks with little deformation. The Rb–Sr wholerock –biotite isochron method suggests an age of c. 455 Ma (Hoch et al. 2001). This age is a few tens of million years younger than the Pan-African II event; however, the magma activity is related to the post-collisional events (Hoch et al. 2001) and the age is consistent with that of the post-kinematic mafic dykes from the Sør Rondane Mountains. Therefore, this paper considers their data to characterize the post-kinematic mafic dyke suite in the Mu¨hlig-Hofmann region. The Sør Rondane Mountains (eastern Dronning Maud Land) consist of metamorphic rocks and various kinds of intrusive rocks. The metamorphic rocks comprise banded gneisses of various compositions with minor amounts of pelitic gneisses (Fig. 2b) (Shiraishi et al. 1997). The main structural features of the metamorphic rocks are controlled by the east– west trend of the foliations and fold axes (Toyoshima et al. 1995). The Main Shear Zone and Sør Rondane Suture are considered
238 M. OWADA ET AL.
Fig. 2. Geological sketch map of the Mu¨hlig-Hofmann region (a) (after Owada et al. 2003) and the Sør Rondane Mountains (b) (after Shiraishi et al. 1997).
GEOCHEMISTRY OF POST-KINEMATIC MAFIC DYKE
to be the dominant tectonic boundaries in the mountains (Shiraishi et al. 1991; Osanai et al. 1992). Meta-tonalite with intermediate to felsic compositions is exposed in the southern part of the Main Shear Zone, whereas metasedimentary rocks and various intrusive rocks crop out in the northern part of the Main Shear Zone (Shiraishi et al. 1991). The Sør Rondane Suture divides the region into granulite-facies metamorphic rocks in the north and amphibolite- to greenschistfacies metamorphic rocks in the south (Osanai et al. 1992). A Rb –Sr whole-rock isochron study revealed that the protolith of the meta-tonalite in the southwestern part of the Sør Rondane Mountains was emplaced in the Neoproterozoic era (c. 960 Ma) (Takahashi et al. 1990). The protolith age of an orthogneiss (charnockitic gneiss) that is located north of the Sør Rondane Suture is c. 1000 Ma, as determined by the Rb– Sr and Sm–Nd wholerock isochron methods (Shiraishi & Kagami 1992). The granulite-facies metamorphism occurred
239
c. 530 Ma, as determined in terms of the monazite and zircon chemical Th –U –total Pb isochron method (CHIME) ages (Asami et al. 2005). The felsic intrusive rocks occurring as stocks in the central part of the Sør Rondane Mountains have been dated to between 530 and 500 Ma by the Rb–Sr and Sm –Nd whole-rock isochron methods (Takahashi et al. 1990; Tainosho et al. 1992; Arakawa et al. 1994). Therefore, the timing of the granulite-facies metamorphism and the felsic magma activities in the Sør Rondane Mountains corresponds to the Pan-African II event age. Mafic dykes (lamprophyre and dolerite) intrude the metamorphic rocks but are cut in places by younger veins of pegmatite (Shiraishi et al. 1988). The radiogenic ages (K –Ar and Ar– Ar whole-rock methods) of the mafic dykes imprecisely indicate an age of 434–488 Ma (Takigami et al. 1987; Takigami & Funaki 1991). The magma activities probably occurred at a few tens of million years after the major orogenic events (Arima & Shiraishi 1993).
Fig. 3. Photomicrographs of dolerite, lamprophyre and lamproite. All photographs were taken with plane-polarized light. (a, b) Lamprophyre from the Sør Rondane Mountains; (c) dolerite from the Sør Rondane Mountains; (d) lamproite from the Mu¨hlig-Hofmann region. Bt, biotite; Hbl, hornblende; Epi, epidote; Cpx, clinopyroxene; Aln, allanite. Scale bars in (a–c) represent c. 0.8 mm; scale bar in (d) represents c. 0.5 mm.
240
M. OWADA ET AL.
Table 2. Whole-rock chemical compositions of mafic dykes Location: SR SR SR SR SR SR SR SR Sample: 091012502A 0910125T-AF MA88012509-3 MA88012308-1 091012504C MA88012905 HM88011108 HM88011002 Lamprophyre Lamprophyre Lamprophyre Lamprophyre Lamprophyre Lamprophyre Dolerite Lamprophyre (wt %) SiO2 TiO2 Al2O3 Fe2OT3 MnO MgO CaO Na2O K2O P2O5 LOI Total ppm Ba Co Cr Hf Nb Ni Pb Rb Sr Ta Th U V Y Zn Zr La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Ce/Sm Sm/Yb
52.79 0.98 12.10 7.89 0.13 10.13 9.32 2.58 2.26 0.34 1.10 99.62
53.38 1.23 14.59 7.74 0.12 7.31 8.04 3.25 2.79 0.39 0.96 99.80
55.98 2.20 13.93 7.88 0.13 4.05 5.54 1.04 6.00 1.42 0.61 98.78
46.82 3.98 11.33 7.08 0.10 8.34 8.05 0.20 6.60 3.89 1.68 98.07
1980 43 480 4.7 7.6 170 27 60 1160 0.41 13.7 2.28 127 26 100 170 75.9 143 16.5 66.9 12.2 3.20 9.24 1.06 5.02 0.86 2.41 0.33 2.03 0.28 11.72 6.01
2450 31 130 5.8 8 60 43 55 1320 0.42 17.4 3.18 128 27 40 206 89.6 169 19.3 75.8 13.4 3.63 10.2 1.15 5.37 0.93 2.60 0.37 2.23 0.32 12.61 6.01
3720 19 70 13.7 23.6 30 15 151 895 1.28 7.52 2.74 110 37 90 1000 79.6 144 16.0 63.2 11.7 3.80 9.76 1.29 6.54 1.20 3.47 0.50 3.20 0.44 12.31 3.66
3990 24 360 7.3 16.6 80 32 153 1980 1.71 1.03 0.42 119 42 140 159 107 226 28.8 122 21.8 5.88 18.2 2.07 9.21 1.50 3.78 0.46 2.44 0.30 10.37 8.93
Occurrence and petrography of the sampled dykes Mafic dykes have sparsely intruded the entire area from central to eastern Dronning Maud Land. Most of the mafic dykes clearly cut the foliation of the basement rocks. There is no systematic orientation of the dyke directions. However, some of the dykes were intruded parallel to the pervasive banding of the host gneisses. The mafic dykes are divided into
53.64 1.79 18.58 7.41 0.06 2.85 4.87 4.37 4.81 0.80 0.75 99.93 3307 na 6 na 46 10 22 155 1334 na 15 na 142 33 88 387 na na na na na na na na na na na na na na
50.43 2.89 16.06 13.40 0.18 3.94 6.24 3.21 2.53 0.60 0.31 99.79 988 na bd na 24 16 na 63 729 na na na 195 44 163 336 na na na na na na na na na na na na na na
54.93 1.77 16.61 7.74 0.11 3.61 5.88 4.04 3.34 0.63 0.60 99.26 2634 na 11 na 17 25 na 94 1477 na na na 129 36 89 315 na na na na na na na na na na na na na na
56.13 1.87 16.10 8.32 0.11 2.49 3.82 3.05 5.56 0.71 0.58 98.74 2551 na bd na 34 11 na 232 1346 na na na 100 34 111 371 na na na na na na na na na na na na na na
three types as follows in terms of their lithology: calc-alkaline lamprophyre (vogesite and minette), dolerite and lamproite. The samples used in this study are lamprophyre, dolerite and lamproite. Lamprophyres are medium-grained and darkcoloured rocks that exhibit a preferred orientation corresponding to that of biotite; that is, parallel to the intrusive direction. They locally exhibit a recrystallized lepidoblastic texture, but the igneous texture is apparently preserved (Fig. 3a and b). The
GEOCHEMISTRY OF POST-KINEMATIC MAFIC DYKE
241
Table 2. Continued SR SR SR SR MH MH MH MH MH HM88011704 HM88012403B MA88011201 MA88012411 02010202C 02010901A 02011002A 02011002B 02011002C Lamprophyre Lamprophyre Dolerite Lamprophyre Lamprophyre Lamprophyre Lamproite Lamproite Lamproite 59.63 2.00 15.65 7.74 0.09 2.29 4.30 3.74 2.18 0.47 1.83 99.92 712 na bd na 31 13 na 90 789 na na na 157 25 135 313 na na na na na na na na na na na na na na
54.77 2.54 15.58 10.11 0.13 3.06 5.65 3.88 2.23 0.67 1.36 99.98 796 na 20 na 37 21 na 53 885 na na na 109 32 138 347 na na na na na na na na na na na na na na
51.77 2.48 17.73 8.99 0.27 3.37 4.83 3.98 3.80 0.73 0.52 98.47 2172 na 9 na 40 23 na 550 981 na na na 127 71 224 344 na na na na na na na na na na na na na na
53.87 2.37 16.66 10.26 0.16 3.35 5.52 4.35 2.17 0.64 1.00 100.35 595 na 21 na 35 21 na 78 849 na na na 126 33 158 349 na na na na na na na na na na na na na na
main minerals are brown biotite, greenish brown hornblende and alkali feldspar, with trace amounts of epidote and apatite. The lamprophyres vary from vogesite to the minette types; most of them contain euhedral to subhedral biotite and hornblende and some samples have biotite phenocrysts (Fig. 3a). Epidote shows subhedral shapes with pale greenish yellow to pale yellow pleochroism (Fig. 3b). Calcite occurs interstitially in the matrix (Fig. 3b). Apatite is a common accessory mineral with euhedral to subhedral shapes. Anhedral biotite and
55.21 1.69 17.86 8.14 0.07 2.86 5.12 2.94 3.87 0.45 1.45 99.67
43.68 3.55 10.59 19.59 0.26 5.29 8.12 1.83 2.11 1.13 0.40 96.55
42.72 2.31 10.17 10.12 0.16 11.25 9.09 0.66 7.24 3.10 0.71 97.53
40.45 2.54 10.91 11.35 0.15 12.16 8.12 0.52 7.87 3.62 0.81 98.50
41.03 2.53 11.01 11.06 0.15 11.85 7.28 0.54 7.88 3.15 0.67 97.17
2470 23 bd 20.6 29 na 36 214 897 1.07 32.6 2.32 88 34 236 511 197 374 41.4 155 24.1 4.53 14.8 1.71 7.22 1.08 2.78 0.32 1.87 0.24 15.48 12.93
1270 28 25 77.4 96 bd 23 76 588 4.63 23.7 3.69 204 195 529 2272 460 1150 132 526 95.7 7.86 65.9 8.77 40.3 6.49 18.4 2.44 13.63 1.86 12.01 7.02
9120 38 474 2.8 26.2 197 9 474 2080 0.69 37.7 4.22 193 103 361 1257 352 694 81.3 327 62.4 15.0 47.2 5.79 23.5 3.20 7.34 0.80 4.85 0.62 11.13 12.86
9040 41 1450 14.2 19.3 208 6 526 1910 0.38 29.8 3.34 196 92 399 858 336 670 77.2 308 57.4 13.9 43.4 5.18 20.9 2.82 6.43 0.66 4.06 0.50 11.68 14.12
9770 41 1450 13.6 19.5 209 5 520 1900 0.43 29.6 3.50 199 94 319 920 320 650 76.4 309 57.6 14.0 43.1 5.27 21.1 2.85 6.57 0.68 4.13 0.51 11.28 13.96
greenish to bluish green hornblende are present in the form of recrystallized minerals (Fig. 3a and b). Dolerites are fine-grained rocks that show igneous texture, with the exception of some locally recrystallized varieties. They generally show intergranular to sub-ophitic textures and are composed of plagioclase, clinopyroxene and biotite (Fig. 3c). Ilmenite, apatite and titanite are common as accessory minerals. Some of the dolerites contain fine-grained quartz xenocrysts surrounded by finegrained clinopyroxene and biotite.
242
M. OWADA ET AL.
Table 2. Continued Location: Sample:
(wt %) SiO2 TiO2 Al2O3 Fe2OT3 MnO MgO CaO Na2O K2O P2O5 LOI Total ppm Ba Co Cr Hf Nb Ni Pb Rb Sr Ta Th U V Y Zn Zr La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu Ce/Sm Sm/Yb
MH 02011002D
MH 02011002E Lamproite
MH 02011002F Lamproite
MH 02011002G Lamproite
MH 02011002H Lamproite
MH 02011002I Lamproite
MH 02011002J Lamproite
MH 02011002K Lamproite
40.27 2.53 10.74 11.54 0.16 12.20 7.81 0.50 7.70 3.57 0.63 97.65
41.01 2.46 10.27 11.29 0.16 12.17 8.49 0.53 7.44 3.48 0.60 97.90
39.90 2.50 9.84 12.81 0.18 13.86 7.58 0.37 7.38 2.81 0.87 98.11
42.46 2.43 9.79 10.65 0.17 11.78 9.42 0.60 7.08 3.22 0.67 98.27
42.53 2.01 9.50 10.10 0.15 13.51 8.75 0.63 6.86 3.10 0.59 97.73
42.37 2.00 9.46 10.06 0.15 13.46 8.72 0.62 6.84 3.09 0.64 97.42
43.43 2.52 11.10 10.32 0.15 10.74 7.88 0.59 7.94 2.96 0.59 98.21
44.96 2.28 11.17 8.71 0.13 9.22 8.31 0.73 7.94 2.97 0.53 96.96
9550 44 1560 15.2 18.6 223 6 555 1970 0.38 28.5 3.26 208 94 386 841 335 672 77.9 311 57.3 13.7 42.8 5.15 20.7 2.82 6.39 0.66 4.00 0.50 11.72 14.33
9340 43 1210 16.4 18.9 235 7 523 1820 0.54 39.6 4.61 203 98 402 1387 342 654 77.7 317 58.2 13.8 44.6 5.41 22.2 2.99 7.02 0.73 4.55 0.59 11.24 12.78
7920 50 1270 22.8 12.2 276 bd 574 1310 0.11 24.3 4.05 222 74 382 901 280 527 62.2 251 46.0 10.9 35.0 4.10 16.7 2.28 5.35 0.56 3.50 0.49 11.46 13.15
9450 39 519 18.7 26.7 210 7 493 1980 0.69 35.6 4.04 200 108 371 1040 335 653 81.2 338 65.4 15.7 49.5 6.15 25.0 3.33 7.73 0.82 4.96 0.64 9.99 13.20
9130 47 664 8.94 30.5 421 bd 443 1790 0.74 26.1 3.82 176 88 221 953 324 612 69.8 279 52.3 12.3 40.1 4.86 19.7 2.67 6.13 0.65 3.87 0.52 11.70 13.49
11400 34 790 5.60 23.0 238 9 457 2690 0.76 34.8 3.92 183 100 195 994 397 737 83.1 332 61.3 14.3 45.7 5.54 22.1 3.01 6.88 0.72 4.27 0.55 12.02 14.34
11200 35 1010 17.4 27.4 160 9 487 1930 0.54 37.7 4.61 196 97 267 1231 302 607 75.3 312 60.6 14.6 47.3 5.70 23.5 3.12 7.29 0.77 4.76 0.59 10.01 12.73
13600 31 661 19.7 37.6 141 13 445 2810 1.05 44.0 5.36 175 110 218 1304 346 647 81.1 336 66.2 15.9 51.4 6.26 25.7 3.47 8.04 0.85 5.34 0.65 9.77 12.40
bd, below detection; na, not analysed. SR, Sør Rondane Mountains; MH, Mu¨hlig-Hofmann region.
A lamproite dyke of 1–2 m thickness exhibits a marked banded structure. This dyke discontinuously intrudes a syenite –charnockite complex. It appears to be a synplutonic dyke. The mineral assemblages of each band are of similar mineral composition but show differences in the modal amounts. We collected 11 samples from each
band used for the geochemical investigation. Most minerals show subhedral to anhedral shapes. The dyke mainly consists of biotite, clinopyroxene, K-feldspar and hornblende (Fig. 3d). Allanite and titanite are also present as interstitial minerals. Furthermore, zircon and apatite are also present as accessory minerals.
GEOCHEMISTRY OF POST-KINEMATIC MAFIC DYKE
Major and trace element geochemistry The major and trace (Rb, Sr, Zr, V, Zn, Cr and Ni) elements were determined using X-ray fluorescence (XRF) spectroscopy at the Center for Instrumental Analyses at Yamaguchi University. Other trace elements, including rare earth elements (REE), were analysed by inductively coupled plasma mass spectrometry (ICP-MS) at the Actlabs Co. Ltd., Canada. The analytical results are listed in Table 2. Most mafic dykes plot within the alkaline field in the total alkali –SiO2 (TAS) diagram; however, they are close to the boundary with the calc-alkaline field (Fig. 4a). Lamproite from the Mu¨hlig-Hofmann region and some of the lamprophyres from the Sør Rondane Mountains show chemical compositions of the ultra-K series, whereas the other mafic dykes plot into the shoshonitic series, as shown in the K2O v. Na2O diagram (Fig. 4b). The variation diagrams of mafic dykes are shown in Figure 5. The SiO2 contents of the mafic dykes range between 40 and 60 wt%. The SiO2 and Al2O3 contents increase with decreasing MgO content, whereas the CaO content shows a broad positive correlation with the MgO content. The TiO2 content varies from 1 to 4 wt% and shows no correlation with the MgO content. Most of the mafic dykes show an enrichment in the Sr content (.500 ppm). Lamproite from the Mu¨hlig-Hofmann region has higher Ni, Cr, MgO and Rb contents as compared with the other rock types. Primitive mantle-normalized diagrams (Sun & McDonough 1989) are shown in Figure 6. All mafic dykes have high concentrations of large ion lithophile elements (LILE) compared with high field strength elements (HFSE). They also show
243
Nb, Ta and Ti depletion and positive Ba anomalies with an abundance greater than Rb in the spider diagram. Chondrite-normalized REE patterns of the mafic dykes indicate highly fractionated light REE to heavy REE ratios (LREE/HREE) (Fig. 7). Lamproites from the Mu¨hlig-Hofmann region show a different geochemical character. The primitive mantle-normalized patterns of lamproites show a high abundance of LILE with marked Nb, Ta and Ti depletions (Fig. 6d) and a higher LREE/HREE ratio (Fig. 7d) as compared with the ratios of lamprophyres and high-K dolerites from the Sør Rondane Mountains. In addition, the lamproites exhibit a marked negative Pb anomaly, whereas the lamprophyres and high-K dolerites may even show positive Pb spikes (Fig. 6a, b and d).
Sr and Nd isotope geochemistry Analytical procedure Isotopic analyses were performed by thermal ionization mass spectrometry (TIMS), using a MAT262 system equipped with nine dynamic Faraday cups at Niigata University and the experimental procedure of Kagami et al. (1987). The 87Sr/86Sr ratios and 143 Nd/144Nd ratios were normalized to 86Sr/ 86 Sr ¼ 0.1194 and 146Nd/144Nd ¼ 0.7219, respectively. The normalized 87Sr/86Sr ratios were corrected using the NBS-987 standard 87Sr/86Sr ¼ 0.710241. The 143Nd/144Nd ratios were corrected with the Japanese standard JNdi-1 ¼ 0.512106, which has been well characterized using the international standard La Jolla value of 143Nd/144Nd ¼ 0.511849 (Tanaka et al. 2000). The Rb and Sr concentrations were determined by XRF. The Sm and Nd concentrations were
Fig. 4. TAS (a) and K2O v. Na2O (b) diagram for lamprophyre, dolerite and lamproite. Alkaline and non-alkaline fields are quoted from Le Maitre (2002) for (a) and compositional fields are taken from Turner et al. (1996) for (b). Data are compiled from this study and Ikeda et al. (1995).
244
M. OWADA ET AL.
Fig. 5. Variation diagrams of major and trace elements for lamprophyre, dolerite and lamproite. Data sources are the same as in Figure 4.
GEOCHEMISTRY OF POST-KINEMATIC MAFIC DYKE
245
Fig. 6. Spider diagrams for lamprophyre (a, c), dolerite (b) and lamproite (d). Data normalized to primitive mantle values (Sun & McDonough 1989). Data sources are the same as in Figure 4.
measured by ICP-MS. The analytical errors for 87 Rb/86Sr and 87Sr/86Sr were 5% (1s) and 0.01% (1s), respectively, and those for 147Sm/144Nd and 143Nd/144Nd ratios were 1% (1s) and 0.01% (1s), respectively. The initial Sr and Nd isotope ratios were calculated using the decay constants l87Rb ¼ 1.42 10211 a21 (Steiger & Ja¨ger 1977) and l147Sm ¼ 6.54 10 – 12 a21 (Lugmair & Marti 1978). The detailed isotopic analytical procedures were reported by Miyazaki & Shuto (1998).
Results The Rb–Sr and Sm–Nd isotopic data are given in Table 3. The initial Nd and Sr ratios are calculated for 500 Ma for the mafic dykes (Fig. 8). The isotopic compositions of the granitic rocks and the
compositional field of orthogneisses from the same areas are plotted in this diagram. The isotopic features of the mafic dykes in the Sør Rondane Mountains indicate relatively enriched Nd isotope values, whereas those from the Mu¨hlig-Hofmann region are characterized by negative 1Ndi (210 to 217) values and moderately enriched Sr isotope values. The isotopic compositions of orthogneisses from the Sør Rondane Mountains and the Mu¨hlig-Hofmann region plot within different fields and resemble those of the mafic dykes and the granitic rocks from each region (Fig. 8). The isotopic signatures of the minettes in the Schirmacher Oasis, which are located in the coastal region of the central Dronning Maud Land (Fig. 8), show similar negative 1Ndi values and moderately enriched Sr isotopic values as compared with the mafic dykes from the Mu¨hlig-
246
M. OWADA ET AL.
Fig. 7. Chondrite-normalized REE patterns for lamprophyre (a, c), dolerite (b) and lamproite (d). Chondrite values are quoted from Sun & McDonough (1989). Data sources are the same as in Figure 4.
Hofmann region (Fig. 8). Consequently, the postkinematic intrusive rocks (mafic dykes and granitic rocks) and the orthogneisses from the Sør Rondane Mountains are isotopically different from those of the Mu¨hlig-Hofmann region.
Discussion Crustal contamination and mantle heterogeneity The patterns of the incompatible elements of the mafic dykes show crustal signatures in their Nb, Ta and Ti negative anomalies (Fig. 6) (Jahn et al. 1999). As described above, some dolerite dykes contain quartz xenocrysts, which suggests crustal contamination. To examine the crustal contamination during magma ascent, we test the geochemical features, including the isotopic characteristics
of the mafic dykes. Figure 9 shows a MgO –Sri (at 500 Ma) diagram in which mafic dykes from the region extending from central to eastern Dronning Maud Land are plotted. Dolerites show increasing Sri values with decreasing MgO content, which suggests crustal contamination. On the other hand, other mafic dykes do not show chemical correlation between the Sri and MgO values. In fact, the two mafic dyke suites possessing low Cr, Ni and Mg contents (mainly dolerites but also some Sør Rondane lamprophyres) show significant evidence of crustal contamination during magma transport and/or emplacement. Other mafic dykes with some ‘crustal contributions’ must have been derived from enriched mantle sources via crustal subduction, as they clearly show primary geochemical characteristics. Their mantle-derived geochemical features are well preserved and no significant crustal contamination is detectable after mantle segregation.
GEOCHEMISTRY OF POST-KINEMATIC MAFIC DYKE
247
Table 3. Isotopic compositions of mafic dykes Sample Sør Rondane O910125T-AF MA88012308-1 Mu¨hlig-Hofmann 02011002F 02011002H 02010901A Sample Sør Rondane O910125T-AF MA88012308-1 Mu¨hlig-Hofmann 02011002F 02011002H 02010901A
Rb/86Sr
87
1320 1980
0.1205 0.2235
0.70726 0.70565
0.70640 0.70405
574 443 76
1310 1790 588
1.2690 0.7165 0.3741
0.71781 0.71434 0.71105
0.70877 0.70924 0.70838
Lithology
Nd (ppm)
Sm (ppm)
Lamprophyre Lamprophyre
75.8 122
13.4 21.8
0.1069 0.1080
Lamproite Lamproite Lamprophyre
251 279 526
46.0 52.3 95.7
0.1108 0.1133 0.1100
Rb (ppm)
Sr (ppm)
Lamprophyre Lamprophyre
55 153
Lamproite Lamproite Lamprophyre
Lithology
The mafic dykes having high MgO content (.7 wt%) show variable Sri values. The lamproites in the Mu¨hlig-Hofmann region show high Sri values and have the highest MgO content among the mafic
87
147
Sm/144Nd
143
Sr/86Sr
Nd/144Nd
Sri (500 Ma)
Ndi (500 Ma)
1Ndi
0.511884 0.512215
0.511705 0.512034
25.65 0.78
0.511511 0.511524 0.511803
0.511325 0.511334 0.511618
213.06 212.89 27.34
dykes in the region extending from central to eastern Dronning Maud Land. These geochemical characteristics indicate mantle heterogeneity and magma derivation from different mantle sources.
Residual phases in the mantle sources The primitive mafic dykes indicate the presence of shoshonitic and ultra-K materials in the K2O v. Na2O diagram (Fig. 4b). In addition, primitive
Fig. 8. 1Nd and 87Sr/86Sr ratios calculated at 500 Ma for dolerite, lamprophyre and lamproite. This diagram also shows isotopic values of granitic rocks and metamorphic rocks from central to eastern Dronning Maud Land and minette from Schirmacher Oasis. Data are quoted from the following sources. Sør Rondane Mountains: mafic dykes: this study and Ikeda et al. (1995); metamorphic rocks: Grew et al. (1992) and Shiraishi & Kagami (1992); granitic rocks: Arakawa et al. (1994). Mu¨hlig-Hofmann region: mafic dykes: this study and Hoch et al. (2001); metamorphic rocks and syenite: Jacobs et al. (2003).
Fig. 9. 87Sr/86Sr ratios (corrected at 500 Ma)– MgO (wt%) diagram for dolerite, lamprophyre and lamproite. Data are from Ikeda et al. (1995) for the dolerite and lamprophyre from the Sør Rondane Mountains, and from this study for the lamprophyre and lamproite from the Mu¨hlig-Hofmann region.
248
M. OWADA ET AL.
mantle-normalized patterns show negative Nb, Ta and Ti anomalies. The geochemical features of these primitive magmas resemble those of subduction-related basaltic magma that probably left Ti-bearing residual phases at the source regions. Figure 10 shows a graph of Mg number (100 Mg/(Mg þ Fe); mole) v. primitive mantlenormalized Rb/Ba and Nb/Ta ratios for the mafic dykes with MgO content greater than 7 wt%. The Mg numbers of these mafic dykes are greater than 60 and mostly higher than 65, indicating the presence of slightly evolved magmas but those that are close to primitive or primary compositions. As described above, these post-kinematic mafic dykes show high K, Rb and Ba contents. In general, a mantle-derived melt produced by low degree of melting has a high concentration of incompatible elements such as K, Rb and Ba. Indeed, the lamprophyre and lamproite with MgO content greater than 7 wt% show high Ce/Sm and Sm/Yb ratios (Table 2), indicating that the melting rates are small (Hawkesworth et al. 1994). The Ba contents of mafic dykes are up to 4000 ppm for the lamprophyre from the Sør Rondane Mountains and 13 600 ppm for the lamproite from the Mu¨hlig-Hofmann region (Table 2). The Ba content of the primitive mantle is c. 7 ppm (Sun & McDonough 1989). The inferred degree of melting is 0.2% for the lamprophyre and 0.05% for the lamproite if the bulk partitioning coefficient of Ba is close to zero. These low degrees of melting are unlikely to cause the segregation of the mafic dykes from the mantle sources. Therefore, the mafic dykes would be derived from a metasomatized mantle with initial enrichment in LILE. The studied mafic dykes show high LILE contents (Fig. 6). Moreover, the Ba primitive mantlenormalized values are higher than those for Rb
(Fig. 6). The Rb/Ba normalized ratios of the mafic dykes are less than 1.0 (Fig. 10a). Phlogopite has higher partition coefficients for Rb than Ba (Green 1994); this suggests that the primitive melts of the mafic dykes are equilibrated with the phlogopite at the source region. The primitive mantle-normalized Nb/Ta ratios of the lamprophyre from the Sør Rondane Mountains are less than 1.5, whereas those of the lamproite from the Mu¨hlig-Hofmann region are greater than 1.7 (Fig. 10b). Rutile and titanite are important Ti phases in the upper mantle. Rutile has higher partition coefficients for Nb as compared with those for Ta, whereas titanite preferably has higher partition coefficients for Ta as compared with those for Nb (Guo et al. 2006). Consequently, the primitive magmas of the mafic dykes in the region extending from central to eastern Dronning Maud Land were derived from the phlogopite-rich metasomatized mantle, leaving rutile as the residual phase. In addition, the lamproite magma also coexists with titanite during melting.
Tectonic setting of the mafic magmatism of the region extending from central to eastern Dronning Maud Land during the Pan-African event High-K rocks such as lamproite suites occur in a wide variety of tectonic settings, including continental rift zones, oceanic islands, island arcs, active continental margins and continental collision zones (Foley et al. 1987; Rock 1987). Mikhalsky (2004) inferred that slightly deformed shoshonitic dykes from central Dronning Maud Land (9– 148E) originated in a late-orogenic compressional environment rather than in a riftogenic extensional
Fig. 10. Mg-number v. Rb/Ba (a) and Nb/Ta (b) ratios normalized to primitive mantle for lamprophyre and lamproite having MgO .7 wt%. Normalized values are taken from Sun & McDonough (1989). Data sources are the same as in Figure 4.
GEOCHEMISTRY OF POST-KINEMATIC MAFIC DYKE
environment, based on geochemical investigations of the major and trace elements. The mafic dykes of Dronning Maud Land studied here show negative Nb, Ta and Ti anomalies similar to those of subduction-related magmas. To determine the tectonic setting for this mafic magmatism, mafic dykes with MgO content greater than 7 wt% are plotted on selected discrimination diagrams (Fig. 11). Lamproites from the Mu¨hlig-Hofmann region plot not only in the field of within-plate magma (Fig. 11a) but also in that of subduction-related magmas (Fig. 11b). On the other hand, lamprophyre from the Sør Rondane Mountains includes both within-plate and arcrelated fields (Fig. 11a and b). These data combined with the spider diagram normalized to the primitive mantle (Fig. 6) suggest that the lamprophyre and lamproite magmas are characterized by geochemical features in both within-plate and arc-related environments. Therefore, the inferred petrogenetic model in this study is consistent with previously reported models; according to these models, the mafic dyke magmas in eastern Dronning Maud Land have been formed in a within-plate tectonic setting by the mixing of subduction-related materials at mantle depths (Arima & Shiraishi 1993; Ikeda et al. 1995; Zhao et al. 1995). The P– T evolution of the metamorphic rocks that occur in the Sør Rondane Mountains shows a clockwise path (Asami et al. 1992). Similar metamorphic evolution has been reported from central Dronning Maud Land (Engvik & Elvevold 2004),
249
which is located c. 700 km west of the Sør Rondane Mountains. In the region extending from central to eastern Dronning Maud Land, the peak metamorphic event occurs at 570 and 530 Ma relative to the Pan-African event, which is generally accepted as the encompassment stage of the continental collision between East and West Gondwana and the subsequent collapse of the orogen (Paech 1997; Jacobs et al. 1998, 2003; Asami et al. 2005). Taking metamorphic processes and age dating into account, the region extending from central to eastern Dronning Maud Land should be located in the collision zone between West and East Gondwana during the Pan-African event. Most of the studied mafic magmatism occurred at a few million to a few tens of million years after the metamorphic peak (Shiraishi & Kagami 1992; Shiraishi et al. 1994). It is reasonable to define this mafic magma activity as a postkinematic igneous activity possibly linked to the extensional stage after the West and East Gondwana collision event. Lamproites from the Mu¨hlig-Hofmann region show contrasting geochemical characteristics such as high K2O contents (Fig. 4b), negative Nb, Ta, Pb and Ti anomalies in the primitive mantlenormalized patterns (Fig. 6d), and higher LREE/ HREE ratios (Fig. 7d) as compared with those of the lamprophyres from the Sør Rondane Mountains (Figs 6a and 7a). Minettes from the Schirmacher Oasis in central Dronning Maud Land, located c. 100 km north of the Mu¨hlig-Hofmann region,
Fig. 11. Lamprophyre (Sør Rondane Mountains) and lamproite (Mu¨hlig-Hofmann region) in discrimination diagrams for basaltic rocks. (a) Zr– Y diagram (after Mu¨ller et al. 1992); (b) Ta/Yb–Th/Yb diagram (after Pearce 1983). Data sources are the same as in Figure 4.
250
M. OWADA ET AL.
exhibit negative Nb and Ti anomalies and negative 1Ndi values with moderately enriched Sri ratios (Hoch et al. 2001). These geochemical signatures, especially Sr and Nd isotope compositions, are similar to those of the studied lamproites (Fig. 8). The syntectonic high-grade metamorphic event could have occurred at c. 570 Ma in the Mu¨hlig-Hofmann region (Jacobs et al. 2003), whereas the timing of the peak metamorphism is considered to be c. 530 Ma in the Sør Rondane Mountains (Asami et al. 2005); however, the magma activities of the post-kinematic intrusive rocks are coeval in both regions. In addition, the Sr and Nd isotope signatures of the mafic dykes are different for central and eastern Dronning Maud Land; those of high-grade basement rocks and granitoids are also different for each area (Fig. 8). As mentioned above, geochemical features involving the isotope signatures of the mafic dykes reflect the composition of the subcontinental lithospheric mantle, which was already metasomatized before the formation of the high-K mafic magmas. The continental crusts that separated from the mantle possess different isotopic signatures as a consequence of their different Rb/Sr and Sm/Nd ratios. The modelled ages for the crustal protoliths are c. 1100 Ma in the Mu¨hlig-Hofmann region and c. 1000 Ma in the Sør Rondane Mountains, suggesting that the metasomatized lithosphere of Dronning Maud Land was produced as a result of a prolonged evolution of at least 500 Ma. In this scenario, the geological histories of the Mu¨hlig-Hofman region and the Sør Rondane Mountains could be different from region to region. According to Dalziel (1991), West and East Gondwana existed up to c. 570 Ma, and the Mozambique Ocean between West and East Gondwana closed in the early Cambrian period. Considering the tectonic setting and geochemical signatures, a large segment boundary such as the collision boundary between West and East Gondwana is assumed to exist between the Mu¨hlig-Hofmann region and the Sør Rondane Mountains. The authors are grateful to K. Shiraishi, M. Asami, T. Toyoshima, T. Kano, T. Kawasaki, H. Ishizuka, T. Yoshimura, T. Miyamoto, A. Kamei and N. Nakano for their valuable discussion, and would like to sincerely thank the members of JARE-32, S. Elvevold, A. La¨ufer and I. Manson for assistance during our fieldwork. Parts of samples using this study were provided by K. Shiraishi and M. Asami. We gratefully acknowledge the editorial suggestions made by M. Satish-Kumar with regard to the manuscript, and thank W. Bauer, C. Vallaseca and R. K. Srivastava for critical reading and improvement of the manuscript. This work was partly supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology, Japan (numbers 15540437 and 18540454 to M.O.; number 17253005 to Y.O.).
References A RAKAWA , Y., T AKAHASHI , Y. & T AINOSHO , Y. 1994. Nd and Sr isotope characteristics of the plutonic rocks in the Sør Rondane Mountains, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geoscience, 7, 49– 59. A RIMA , M. & S HIRAISHI , K. 1993. Geochemical characteristics of metamorphosed high K/Na dykes in eastern Queen Maud Land, Antarctica: Ultrapotassic igneous activity linked to Pan-African orogeny. Proceedings of NIPR Symposium on Antarctic Geoscience, 6, 103–115. A SAMI , M., O SANAI , Y., S HIRAISHI , K. & M AKIMOTO , H. 1992. Metamorphic evolution of the Sør Rondane Mountains, East Antarctica. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Science. Terra, Tokyo, 7 –15. A SAMI , M., S UZUKI , K. & G REW , E. S. 2005. Monazite and zircon dating by the chemical Th– U–total Pb isochron method (CHIME) from the Sør Rondane Mountains, East Antarctica: A reconnaissance study of the Mozambique suture in eastern Queen Maud land. Journal of Geology, 113, 59–82. A USTRHEIM , H., E LVEVOLD , S., E NGVIK , A. K. & P AULSSON , O. 1997. Geological observations in Gjelsvikfjella, Mu¨hlig-Hofmannfjella and western Orvinfjella during NARE 96/97. In: W INTHER , J. G. (ed.) Report of the Norwegian Antarctic Research Expedition 1996/1997. Norsk Polarinstitutts Meddelelser, 148, 77–84. B AUER , W., T HOMAS , R. J. & J ACOBS , J. 2003. Proterozoic–Cambrian history of Dronning Maud Land in the context of Gondwana assembly. In: Y OSHIDA , M., W INDLEY , B. F. & D ASGUPTA , S. (eds) Proterozoic East Gondwana: Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 247–269. B OGER , S. D., W ILSON , C. J. L. & F ANNING , C. M. 2001. Early Paleozoic tectonism within the East Antarctic craton: the final suture between East and West Gondwana? Geology, 29, 463– 466. D ALLMANN , W. K., A USTRHEIM , H., B UCHER N URMINEN , K. & O HTA , Y. 1990. Geology around the Norwegian Antarctic station ‘Troll’, Jutulsessen, Dronning Maud Land. Norsk Polarinstitutt Meddelelser, 111. D ALZIEL , I. 1991. Pacific margins of Laurentia and East Antarctica–Australia as a conjugate rift pair: evidence and implications for an Eocambrian supercontinent Geology, 19, 598– 601. E NGVIK , A. K. & E LVEVOLD , S. 2004. Pan-African extension and near-isothermal exhumation of a granulite facies terrain, Dronning Maud Land, Antarctica. Geological Magazine, 141, 649–660. F ITZSIMONS , I. C. W. 2000. Grenville-age basement provinces in East Antarctica; evidence for three separate collisional orogens. Geology, 28, 879–882. F OLEY , S. F., V ENTURELLI , G., G REEN , D. H. & T OSCANI , L. 1987. The ultrapotassic rocks: Characteristics, classification and constraints for petrogenetic models. Earth-Science Reviews, 24, 81– 134. G REEN , T. H. 1994. Experimental studies of trace-element partitioning applicable to igneous petrogenesis— Sedona 16 years later. Chemical Geology, 117, 1 –36.
GEOCHEMISTRY OF POST-KINEMATIC MAFIC DYKE G REW , E. S., M ANTON , W. I., A SAMI , A. & M AKIMOTO , H. 1992. Reconnaissance geochronologic data on Proterozoic polymetamorphic rocks of the eastern Sør Rondane Mountains, East Antarctica. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Science. Terra, Tokyo, 37–44. G UO , Z., W ILSON , M., L IU , J. & M AO , Q. 2006. Post-collisional, potassic and ultrapotassic magmatism of the northern Tibetan Plateau: Constraints on characteristics of the mantle source, geodynamic setting and uplift mechanisms. Journal of Petrology, 47, 1177–1220. H AWKESWORTH , C. J., G ALLAGHER , K., H ERGT , J. M. & M C D ERMOTT , F. 1994. Destructive plate margin magmatism: Geochemistry and melt generation. Lithos, 33, 169–188. H OCH , M., R EHKA¨ MPER , M. & T OBSCHALL , H. J. 2001. Sr, Nd, Pb and O isotopes of minettes from Schirmacher Oasis, East Antarctica: a case of mantle metasomatism involving subducted continental material. Journal of Petrology, 42, 1387– 1400. I KEDA , Y., S HIRAISHI , K. & K AGAMI , H. 1995. Geochemical characteristics of mafic dike rocks from the Sør Rondane Mountains, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 8, 49– 64. J ACOBS , J. & B AUER , W. 2001. Gjelsvikfjella and Mu¨hlig-Hofmann-Gebirge (E-Antarctica): Another piece of the East Antarctic Orogen? Zeitschrift der Deutschen Geologischen Gesellschaft, 152, 249– 259. J ACOBS , J., F ANNING , C. M., H ENJES -K UNST , F., O LESCH , M. & P AECH , H. J. 1998. Continuation of the Mozambique Belt into East Antarctica: Grenville-age metamorphism and polyphase PanAfrican high-grade events in central Dronning Maud Land. Journal of Geology, 106, 385–406. J ACOBS , J., B AUER , W. & F ANNING , C. M. 2003. Late Neoproterozoic/Early Palaeozoic events in central Dronning Maud Land and significance for the southern extension of the East African Orogen into East Antarctica. Precambrian Research, 126, 27–53. J AHN , B. M., W U , F., L O , C. H. & T SAI , C. H. 1999. Crust– mantle interaction induced by deep subduction of the continental crust: geochemical and Sr –Nd isotopic evidence from post-collisional mafic– ultramafic intrusions of the northern Dabie complex, central China. Chemical Geology, 157, 119–146. K AGAMI , H., I WATA , M., S ANO , S. & H ONMA , H. 1987. Sr and Nd isotopic compositions and Rb, Sr, Sm and Nd concentrations of standard samples. Technical Reports, Institute for Study of the Earth’s Interior, Okayama University, Series 4, 1– 16. L E M AITRE , R. W. (ed.) 2002. Igneous Rocks: A Classification and Glossary of Terms. Cambrigde University Press, Cambridge. L UGMAIR , G. W. & M ARTI , K. 1978. Lunar initial 143 Nd/144Nd: differential evolution of the lunar crust and mantle. Earth and Planetary Science Letters, 39, 349–357. M IKHALSKY , E. 2004. Geochemistry of deformed mafic to intermediate dykes in central Dronning Maud Land, East Antarctica. Geologisches Jahrbuch, B96, 233–253. M IYAZAKI , T. & S HUTO , K. 1998. Sr and Nd isotope ratios of twelve GSJ rock reference samples. Geochemical Journal, 32, 345–350.
251
M OYES , A. B. 1993. The age and origin of the Jutulsessen granitic gneiss, Gjelsvikfjella, Dronning Maud Land. South African Journal of Antarctic Research, 23, 25–32. M U¨ LLER , D., R OCK , N. S. M. & G ROVES , D. I. 1992. Geochemical discrimination between shoshonitic and potassic volcanic rocks in different tectonic settings: A pilot study. Mineralogy and Petrology, 46, 259– 289. O HTA , Y. 1999. Nature environment map, Gjelsvikfjella and Western Mu¨hlig-Hofmannfjella, sheets 1 and 2, Dronning Maud Land. Norsk Polarinstitutt, Temakart nr. 24. O HTA , Y., T ØRUDBAKKEN , B. & S HIRAISHI , K. 1990. Geology of Gjelsvikfjella and western Mu¨hligHofmannfjella, western Dronning Maud Land, East Antarctica. Polar Research, 8, 99– 126. O SANAI , Y., S HIRAISHI , K., T AKAHASHI , Y. ET AL . 1992. Geochemical characteristics of metamorphic rocks from the central Sør Rondane Mountains, East Antarctica. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Science. Terra, Tokyo, 17–27. O WADA , M., B ABA , S., L A¨ UFER , A. L., E LVEVOLD , S., S HIRAISHI , K. & J ACOBS , J. 2003. Geology of Mu¨hlig-Hofmannfjella and Filchnerfjella in Dronning Maud Land, East Antarctica: A preliminary report on a Japan–Norway–Germany joint geological investigation. Polar Geoscience, 16, 108–136. P AECH , H. J. 1997. Pervasive Pan-African reactivation of the Grenvillian crust and large igneous intrusions in central Dronning Maud Land, East Antarctica. In: M ILLER , J. A., H OLDSWORTH , R. E., B UICK , I. S. & H AND , M. (eds) Continental Reactivation and Reworking. Geological Society, London, Special Publications, 184, 343–355. P AULSSON , O. & A USTRHEIM , H. 2003. A geochronological and geochemical sudy of rocks from Gjelsvikfjella, Dronning Maud Land, Antarctica— implications for Mesoproterozoic correlations and assembly of Gondwana. Precambrian Research, 125, 113– 138. P EARCE , J. A. 1983. The role of sub-continental lithosphere in magma genesis at destructive plate margins. In: H AWKESWORTH , C. J. & N ORRY , M. J. (eds) Continental Basalts and Mantle Xenoliths. Shiva, Nantwich, 230–249. P ETERSON , T. D., V AN B REEMEN , O., S ANDEMAN , H. & C OUSENS , B. 2002. Proterozoic (1.85– 1.75 Ga) igneous suites of the Western Churchill Province: granitoid and ultrapotassic magmatism in a reworked Archean hinterland. Precambrian Research, 119, 73–100. R AVICH , M. G. & K RYLOV , A. Y. 1964. Absolute ages of rocks from East Antarctica. In: A DIE , R. J. (ed.) Antarctic Geology. North Holland, Amsterdam, 590– 596. R OCK , N. M. S. 1987. The nature and origin of lamprophyre: An overview. In: F ITTON , J. G. & U PTON , B. G. J. (eds) Alkaline Igneous Rocks. Blackwell, Oxford, 191 –226. S HIRAISHI , K. & K AGAMI , H. 1992. Sm–Nd and Rb–Sr ages of metamorphic rocks from the Sør Rondane Mountains, East Antarctica. In: Y OSHIDA , Y., K AMINUMA ,
252
M. OWADA ET AL.
K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Science. Terra, Tokyo, 29–35. S HIRAISHI , K., K ANISAWA , S. & I SHIKAWA , K. 1988. Geochemistry of post-orogenic mafic dike rocks from Eastern Queen Maud Land, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 2, 117 –132. S HIRAISHI , K., A SAMI , M., I SHIZUKA , H. ET AL . 1991. Geology and metamorphism of the Sør Rondane Mountains, East Antarctica. In: T HOMSON , M. R. A., C RAME , J. A. & T HOMSON , J. W. (eds) Geological Evolution of Antarctica. Cambridge University Press, Cambridge, 77–82. S HIRAISHI , K., E LLIS , D. J., H IROI , Y., F ANNING , C. M., M OTOYOSHI , Y. & N AKAI , Y. 1994. Cambrian orogenic belt in East Antarctica and Sri Lanka: Implications for Gondwana assembly. Journal of Geology, 102, 47–65. S HIRAISHI , K., O SANAI , Y., I SHIZUKA , H. & A SAMI , M. 1997. Geological map of the Sør Rondane Mountains, Antarctica. Antarctica Geological Map Series, sheet 35, scale 1 : 25 0000. National Institute of Polar Research, Tokyo. S TEIGER , R. H. & J A¨ GER , E. 1977. Subcommision of geochronology: convention on the use of decay constants in geo- and cosmochronology. Earth and Planetary Science Letters, 36, 359–362. S TERN , R. J. 1994. Arc assembly and continental collision in the Neoproterozoic East African Orogen: Implications for the consolidation of Gondwana. Earth and Planetary Science Review, 22, 319–351. S UN , S. S. & M C D ONOUGH , W. F. 1989. Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. In: S AUNDERS , A. D. & N ORRY , M. J. (eds) Magmatism in the Ocean Basins. Geological Society, London, Special Publications, 42, 313 –345. T AINOSHO , Y., T AKAHASHI , Y., A RAKAWA , Y., O SANAI , Y., T SUCHIYA , N., S AKIYAMA , T. & O WADA , M. 1992. Petrochemical character and
Rb–Sr isotopic investigation of the granitic rocks from the Sør Rondane Mountains, East Antarctica. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Science. Terra, Tokyo, 45–54. T AKAHASHI , Y., A RAKAWA , Y., S AKIYAMA , T., O SANAI , Y. & M AKIMOTO , H. 1990. Rb–Sr and K– Ar whole rock ages of the plutonic bodies from the Sør Rondane Mountains, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 4, 1–8. T AKIGAMI , Y. & F UNAKI , M. 1991. 40Ar/39Ar and K –Ar ages for igneous and metamorphic rocks from the Sør Rondane Mountains, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 5, 122–135. T AKIGAMI , Y., K ANEOKA , I. & F UNAKI , M. 1987. Age and paleomagnetic studies for intrusive and metamorphic rocks from the Sør Rondane Mountains, Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 1, 169– 177. T ANAKA , T., T OGASHI , S., K AMIOKA , H. ET AL . 2000. JNdi-1: a neodymium isotopic reference in consistency with La Jolla neodymium. Chemical Geology, 168, 279–281. T OYOSHIMA , T., O WADA , M. & S HIRAISHI , K. 1995. Structural evolution of metamorphic and igneous rocks from the central part of the Sør Rondane Mountains, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 8, 75–97. T URNER , S., A RNAUD , N., L IU , J. ET AL . 1996. Postcollision, shoshonitic volcanism on the Tibetan Plateau: Implications for convective thinning of the lithosphere and the source of ocean island basalts. Journal of Petrology, 37, 45–71. Z HAO , J. X., S HIRAISHI , K., E LLIS , D. J. & S HERATON , J. W. 1995. Geochemical and isotopic studies of syenites from the Yamato Mountains, East Antarctica: Implications for the origin of syenitic magmas. Geochimica et Cosmochimica Acta, 59, 1363– 1382.
Geodynamic evolution of Mt. Riiser-Larsen, Napier Complex, East Antarctica, with reference to the UHT mineral associations and their reaction relations TOMOKAZU HOKADA1,2, YOICHI MOTOYOSHI1,2, SATOKO SUZUKI3, MASAHIRO ISHIKAWA4 & HIDEO ISHIZUKA5 1
National Institute of Polar Research, 1-9-10 Kaga, Itabashi, Tokyo 173-8515, Japan (e-mail:
[email protected])
2
The Graduate University for Advanced Studies, 1-9-10 Kaga, Itabashi, Tokyo 173-8515, Japan 3
Graduate School of Science and Technology, Niigata University, Niigata, 950-2181, Japan 4
Environment and Information Science, Yokohama National University, 79-7 Tokiwadai, Hodogaya-ku, Yokohama 240-8501, Japan
5
Department of Geology, Kochi University, 2-5-1 Akebono-cho, Kochi 780-8520, Japan Abstract: Mt. Riiser-Larsen is the largest outcrop in the Archaean–early Proterozoic Napier Complex, East Antarctica. The area is structurally divided into the Main and the Western Blocks by the subvertical Riiser-Larsen Main Shear Zone (RLMSZ) of about 200 m width composed of mylonite and pseudotachylite. Mineral parageneses including sapphirine þ quartz and osumilite, diagnostic of ultrahigh-temperature (UHT) metamorphism, are found in Mg-rich aluminous, siliceous and quartzo-feldspathic gneiss layers in both the Main and the Western Blocks of the Mt. Riiser-Larsen area. Some of the sapphirine–quartz associations are accompanied by retrograde reaction textures, which include growth of cordierite and/or garnet between sapphirine and quartz in the Main Block, and of orthopyroxene þ sillimanite in the Western Block. These textures indicate the reaction sapphirine þ quartz (þ orthopyroxene) ¼ cordierite
(1)
sapphirine þ quartz (þ orthopyroxene) ¼ garnet
(2)
sapphirine þ quartz ¼ orthopyroxene þ sillimanite
(3)
and
in the Main Block and
in the Western Block. Phase equilibria and P –T pseudosections for sapphirine þ quartz-bearing associations suggest that these three reactions took place during a temperature drop from 1100 8C to 1000 8C at pressures of 0.6–0.8 GPa in the Main Block and 0.8–0.9 GPa in the Western Block. The geological structure and distribution of the UHT rocks provide an insight into the vertical extent of the .1000 8C UHT metamorphic zone: a minimum thickness of 4– 5 km of the UHT-metamorphosed layers, which become deeper towards the west in the Main Block. The Western Block represents a c. 0.1–0.3 GPa (c. 3 –10 km) deeper structural level than the Main Block. In addition to the extent of the horizontal distribution of UHT metamorphism in the Napier Complex, our results on the vertical component provide new constraints for modelling the heat source and tectonic process of the unusually high-temperature regional metamorphism in the late Archaean–early Proterozoic. Electron microprobe monazite U– Th– Pb dating for hydrated and mylonitized sapphirine– quartz gneiss gave a wide spectrum of monazite age distribution between 2300 and 800 Ma, suggesting the tectonic uplift and juxtaposition of the two blocks in the Mt. Riiser-Larsen area later than the mid–late Proterozoic.
The Napier Complex is one of Archaean cratonic terranes in the East Antarctic continent. Among the Archaean to Palaeoproterozoic terranes (e.g. Grunehogna, southern Prince Charles Mountains,
Vestfold Hills, Terre Ade´lie, Miller Range and Shackleton Range; inset of Fig. 1) in East Antarctica, the Napier Complex preserves several unique geological and metamorphic features. First, early
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 253 –282. DOI: 10.1144/SP308.13 0305-8719/08/$15.00 # The Geological Society of London 2008.
254
T. HOKADA ET AL.
Fig. 1. Map showing the location of Mt. Riiser-Larsen and the other outcrops exposed on the Antarctic ice sheet. The estimated boundary between the Napier Complex and the Rayner Complex is after Sheraton et al. (1987). Isograd of ‘sapphirine– quartz-in’ is cited from Harley & Hensen (1990). Inset: map of East Antarctica in a reconstruction of Gondwana modified after Fitzsimons (2000) and Harley (2003).
Archaean .3850 Ma protolith ages have been obtained from tonalitic orthogneisses (Black et al. 1986; Harley & Black 1997; Kelly & Harley 2005), which are the oldest rocks in Antarctica and close to the age of the Earth’s oldest rocks, the 4000 Ma Acasta Gneiss in Canada (Bowring et al. 1989). In addition, most of the lithologies in the Napier Complex were formed before the 2590–2450 Ma main tectonometamorphic event(s) (Harley et al. 2001; Kelly & Harley 2005), and thus are critical for understanding the Archaean crustal evolution of the Antarctic continent. Second, the Napier Complex has experienced unusually hightemperature, .900–1100 8C (namely ‘ultrahightemperature (UHT)’) metamorphism on a regional scale; the UHT area covers 200 100 km2 defined by the distribution of diagnostic mineral parageneses including sapphirine þ quartz, orthopyroxene þ
sillimanite þ quartz and osumilite (Harley & Hensen 1990) (Fig. 1). This wide distribution of UHT metamorphism with estimated peak metamorphic temperatures .1120–1150 8C at a relatively shallow crustal depth of 20–30 km (Harley & Motoyoshi 2000; Ishizuka et al. 2002; Harley 2004) requires a unique tectonic model to reasonably explain the unusually high-T conditions achieved at such depths. Another significance of the Napier Complex is its geographical position in Gondwana and pre-Gondwana configurations. The Napier Complex probably formed a single cratonic block earlier than 2850 Ma (e.g. Kelly & Harley 2005), and subsequently evolved through several tectonothermal events, which culminated in the major UHT metamorphism at 2550–2480 Ma (Table 1). At present, it is believed that the Napier Complex had been a part of the Greater India Block
GEOLOGY OF MT. RIISER-LARSEN
255
Table 1. Summary of the proposed ages and events in the Napier Complex Age (Ma) .3850 –3800 .3270 .3000 3000 –2900 2850 –2840 .2830 .2700 2630 2550 –2480 2450 .2400 2400 –2350 2200 2000 –1900 1900 –1500 1200 1000 500
Event(s)
Location
Tonalite magmatism Tonalite magmatism Deposition of supracrustal sequences Various felsic and mafic magmatism High-grade metamorphism (pre-D1) Granite magmatism Deposition of supracrustal sequences Tonalite magmatism High-grade metamorphism (D1 –D2/UHT) Tectonothermal event (D3?) Late granite magmatism Intrusion of mafic dykes, local thermal event? Local thermal event? Intrusion of mafic dykes Intrusion of pegmatite, local fluid activity Intrusion of mafic dykes Intrusion of pegmatites Intrusion of pegmatite and alkali dykes
Mt. Sones, Gage Ridge, (Fyfe Hills?) Mt. Riiser-Larsen, (Tonagh island?) (Various localities) Regional event Napier Mountains, Mt. Riiser-Larsen Tonagh island Regional event Regional event? Napier Mountains (Various localities) Various localities Mt. Riiser-Larsen (Various localities) (Various localities) (Various localities) (Various localities)
Ages and events are modified after the compilations by Sheraton et al. (1987), Harley & Black (1997), Grew (1998), Kelly & Harley (2005), Miyamoto et al. (2006) and Suzuki et al. (2006).
(denoted the India –Antarctic Sector or Rayner Province) in pre-Gondwana reconstructions, and had been separated from the other Archaean– early Proterozoic terranes of East Antarctica (e.g. Fitzsimons 2000; Harley 2003). The post-Archaean events of the Napier Complex, which include minor dykes and pegmatitic intrusions, local rehydration, or mylonite or pseudotachylite shear zones during the Proterozoic and later periods, are of potential importance for Gondwana and pre-Gondwana evolution. Several other aspects on the evolution of the Napier Complex have been discussed by Ishizuka (2008). The largest outcrop (13 6 km2) in the Napier Complex is Mt. Riiser-Larsen, which is located on the northeastern coast of Amundsen Bay (Figs 1 and 2) and preserves a typical and well-established Archaean crustal history. The earliest geological record of Mt. Riiser-Larsen started with .3270 Ma tonalitic magmatism, followed by felsic and mafic magmatism and the deposition of supracrustal rocks during 3000–2600 Ma (Table 2; see references therein). Apparently the tonalitic (.3270 Ma) orthogneiss unit overlies 3000–2600 Ma layered sequences; some minor tonalite layers are interleaved in the layered gneiss units (Fig. 2) as a result of intense deformation during the late Archaean–early Proterozoic metamorphic event. The combination of a relatively massive orthogneiss unit and a layered gneiss unit composed of orthogneiss and paragneiss, observed in the Mt. Riiser-Larsen area, is the typical geological feature of the Napier Complex (e.g. Sheraton et al. 1987), but the mutual chronological
relations of these have not always been established in localities other than Mt. Riiser-Larsen. The 2480 Ma sensitive high-resolution ion microprobe (SHRIMP) age (Hokada et al. 2004; Suzuki et al. 2006) constrains the timing of pervasive zircon growth or recrystallization reflecting the major UHT metamorphism, but the precise timing of the peak UHT event is still in debate (e.g. Kelly & Harley 2005). Later Proterozoic mafic dykes (c. 2000– 1900 Ma and c. 1200 Ma; Suzuki et al. 2008) intruded these Archaean gneisses and shear zones that cut across the geological structures. Relationships between the late Archaean–early Proterozoic metamorphic thermal structure and its modification by later shear zone activity in this area have the potential to contribute to the understanding of the development of India–Antarctic Sector of Gondwana. Mt. Riiser-Larsen has been a key area also for the UHT metamorphic processes. UHT metamorphism is defined as crustal metamorphism that has occurred at peak conditions .900 8C at mid- to deep-crustal levels (Harley 1998). The Archaean Napier Complex is one of the best examples of UHT metamorphism, which is characterized by the widespread occurrence of sapphirine þ quartz, orthopyroxene þ sillimanite þ quartz, osumilite and inverted metamorphic pigeonite (e.g. Dallwitz 1968; Ellis et al. 1980; Grew 1980, 1982; Motoyoshi & Matsueda 1984; Sandiford & Powell 1986, 1988; Harley 1987; Motoyoshi & Hensen 1989; Hensen & Motoyoshi 1992). These mineral parageneses have also been reported from many other UHT terranes or localities, lower crustal xenoliths and contact
256
T. HOKADA ET AL.
Fig. 2. Simplified geological map and cross-section of the Mt. Riiser-Larsen area modified after Ishizuka et al. (1998) and Ishikawa et al. (2000). (a) Sample localities of the sapphirine–quartz-bearing gneisses described in the text. Original sample numbers are prefixed by ‘TH96’ for six-digit samples (i.e. 122709 to be TH96122709) and by TH970 for five-digit ones (i.e. 10610 to be TH97010610). (b) The localities of the samples for which isotopic age data are reported in the literature. (See details in Table 2.) (c) Peak metamorphic temperatures estimated by aluminium content in orthopyroxene coexisting with sapphirine and quartz (*Harley & Motoyoshi 2000) and by feldspar thermometry (**Hokada 2001; ***Hokada & Suzuki 2006). Also shown are the localities of the coexisting orthopyroxene þ sillimanite along with sapphirine þ quartz reported by Grew (1982).
aureoles of anorthosite or gabbro plutons (e.g. Harley 1998, 2004; and references therein). However, the areal distribution of sapphirine þ quartz and osumilite, estimated to be c. 200 100 km2 for the Napier Complex (Harley & Hensen 1990) is exceptionally large compared with other UHT
terranes, where it is either sporadic or restricted to a limited area. The UHT metamorphic rocks including sapphirine þ quartz, orthopyroxene þ sillimanite þ quartz, osumilite and inverted metamorphic pigeonite in the Mt. Riiser-Larsen area have been
Table 2. Summary of the ages reported for rocks in Mt. Riiser-Larsen, Napier Complex Rock type
Method
Age*
Ref.
Tonalitic orthogneiss Tonalitic orthogneiss Tonalitic orthogneiss Tonalitic orthogneiss Mafic granulite Opx–bg qtz–fsp gneiss
TH97011302 TH97012401 TH97012816 – – MA88022102 TH97010813 TH97021326 TH97020713 R98022302C
Granitic gneiss
SS97021307
3270 + 12(i), 2789 + 35 (m), 2479 + 15(m) 3267 + 5(i), 2474 + 12(m) c. 3350(i), 2849 + 9(m), 2516 + 8(m), 2451 + 21(m) 3015 + 44 2912 + 186 3080(d), 2750(d), 2436 + 17(m) 2404 + 54(m), 2268 + 17(m), 2034 + 39(m) c. 2900 – 2700(d), 2490 + 107(m) c. 3000 – 2700(d), 2499 + 89(m) c. 3100 – 2700(d), 2452 + 107(m) c. 3000 – 2700(d), 2475 + 84(m) 2488 + 39(m) c. 2830(i), 2514 + 12(m), 2488 + 8(m), 2450 + 4(m) 2476 + 3(m) 2382 + 35(r) 2365 + 22(r) 2380 + 18(r) 2630 + 143 2517 + 5(m), 2480 + 3(m), 2464 + 3(m) 2482 + 2(m) 2470 + 30(m), c. 1900 – 1700(r), c. 700(r) 2445 + 31(m), c. 2300 – 1900(r) 2448 + 129(m) 2420 + 20(m) 2204 + 19(r) 1979 + 80 2078 + 104 c. 1200(i)
1 1 1 2, 3 2, 3 4, 5
Spr –opx –qtz–pl gneiss Spr –opx –os –qtz gneiss Spr –opx –qtz-pl gneiss Grt –opx–os –qtz gneiss
SHRIMP U – Pb Zrn SHRIMP U – Pb Zrn SHRIMP U – Pb Zrn Sm– Nd whole-rock Sm– Nd whole-rock EMP U – Th – Pb Zrn EMP U – Th – Pb Mnz EMP U – Th – Pb Zrn EMP U – Th – Pb Zrn EMP U – Th – Pb Zrn EMP U – Th – Pb Zrn EMP U – Th – Pb Mnz SHRIMP U – Pb Zrn SHRIMP U – Pb Mnz Sm– Nd internal Sm– Nd internal Sm– Nd internal Sm– Nd whole-rock SHRIMP U – Pb Zrn SHRIMP U – Pb Zrn EMP U – Th – Pb Mnz EMP U – Th – Pb Mnz EMP U – Th – Pb Per EMP U – Th – Pb Zrn Sm– Nd internal Rb– Sr whole-rock Sm– Nd whole-rock Rb– Sr whole-rock
Psammitic gneiss Mafic granulite Ultramafic granulite Grt –opx–qtz –fsp gneiss Grt leucosome Altered grt–sil gneiss Spr –opx –os –qtz gneiss
SS97021208-1 SS97021303B – R98022301A R98022301B TH97021409 TH97011607
Spr –bg quartzose gneiss Spr –bg qtz –fsp gneiss Mafic dyke (NE –SW)
MA88021905 SS96122803B –
Mafic dyke (N –S)
–
6 6 6 6 7 7 7 2, 3 6 6 8 9
GEOLOGY OF MT. RIISER-LARSEN
Sample no.
5 10 11 11 11
*Interpretation of age: i, igneous (protolith) age; d, detrital (inherited) age; m, metamorphic age; r, isotopic resetting and closure T age. References: 1, Hokada et al. (2003); 2, Suzuki (2000); 3, Ishikawa et al. (2000); 4, Asami et al. (1998); 5, Asami et al. (2002); 6, Hokada et al. (2004); 7, Suzuki et al. (2006); 8, Hokada & Motoyoshi (2006); 9, Hokada (2007); 10, Suzuki et al. (2001); 11, Suzuki et al. (2008).
257
258
T. HOKADA ET AL.
reported in many earlier studies (e.g. Grew 1982; Harley 1983, 1985; Motoyoshi & Matsueda 1984; Makimoto et al. 1989; Motoyoshi & Hensen 1989; Hensen & Motoyoshi 1992; Ishizuka et al. 1998; Hokada 1999; Harley & Motoyoshi 2000; Suzuki 2000; Ishikawa et al. 2000; Ishizuka et al. 2002; Suzuki et al. 2001; Hokada et al. 2004; Hokada 2007). In particular, Motoyoshi & Hensen (1989) demonstrated that the symplectitic sapphirine– quartz texture formed after cordierite decomposition is suggestive of a counterclockwise prograde P–T path. Harley & Motoyoshi (2000) estimated peak metamorphic temperatures of .1120–1150 8C from the highly aluminous (Al2O3 12 wt%) orthopyroxene coexisting with sapphirine þ quartz in a sample collected from the western margin of this area. A similar metamorphic temperature of c. 1130 8C has been estimated by pyroxene geothermometry applied to ultramafic granulites by Ishizuka et al. (2002). Thus, the Mt. Riiser-Larsen area is a good study area for assessing the UHT metamorphic processes. In this study, we report the coexistence of sapphirine and quartz at 12 localities in the Mt. Riiser-Larsen area (Fig. 2), some of which include reaction textures capable of indicating the cooling path from the peak UHT conditions. A large variety of reaction textures, most of them being formed during the post-UHT cooling stage, have been reported from the metamorphic rocks in the Napier Complex, some of which are indicative of near-isobaric cooling (e.g. Harley & Hensen 1990; Harley 1998). Harley (1998) has suggested that the reaction textures of sapphirine þ quartz rimmed by coronas involving a combination of cordierite, sillimanite and garnet in the northern Napier Complex represent lower pressures during the cooling stage than those of sapphirine þ quartz rimmed by orthopyroxene þ sillimanite in the areas south of the Amundsen Bay region, which is estimated to be the higher-pressure part of the complex. These two different types of breakdown reaction textures from the stable sapphirine þ quartz assemblage have been found in the shear zone dividing two blocks within the Mt. RiiserLarsen area. Based on the indicators of the peak and post-peak metamorphic evolution of the area, we can obtain some insight into the regional tectonic processes.
Geology of Mt. Riiser-Larsen The Napier Complex occupies an area of 400 200 km2 in Enderby Land, East Antarctica (Fig. 1). The Neoproterozoic (c. 900 Ma) granulite terrane of the Rayner Complex bounds the southern and eastern margins of the Napier Complex. Some areas (e.g. Stillwell Hills, Oygarden Islands and
Edward VIII Gulf area) on the Kemp Coast of the Rayner Complex are regarded as a reworked part and hence the eastern extension of the Napier Complex (e.g. Kelly et al. 2005). To the west of the Napier Complex, amphibolite- to granulitefacies, partly UHT, metamorphic belts of early Cambrian age (550–520 Ma) are developed throughout Dronning Maud Land (Forefinger Point, Mt. Vechernyaya, Lu¨tzow-Holm Complex, Yamato–Belgica Complex, Sør Rondane Mountains through central to western Dronning Maud Land; details have been given by Shiraishi et al. 2008). The nature of the boundary between the Napier Complex and surrounding Proterozoic and Cambrian metamorphic belts is not known because it is covered by ice. The distribution of mid-Proterozoic unmetamorphosed dolerite dykes, the ‘Amundsen Dykes’ of Sheraton et al. (1987), defines the area of the Archaean Napier Complex, whereas their metamorphosed equivalents are seen in the surrounding area, which defines late Proterozoic to Cambrian terranes. Figure 2 shows a simplified geological map of the Mt. Riiser-Larsen area (modified after Ishizuka et al. (1998) and Ishikawa et al. (2000)). The eastern part of the area consists of a relatively massive and thick (.200 m) felsic orthogneiss unit, which apparently overlies the layered gneisses (Fig. 3a). The rest of the area is dominated by layered sequences composed mainly of orthopyroxene-bearing felsic orthogneiss (layered gneiss-1; Fig. 3a and b), garnet-bearing felsic gneiss (layered gneiss-2; Fig. 3c and d), and the alternation of various lithologies including both orthogneisses and paragneisses (layered gneiss-3; Fig. 3c and f ). Felsic orthogneisses are the most abundant rock type throughout the study area, and are composed of quartz, feldspar (mesoperthitic alkali feldspar or antiperthitic plagioclase), orthopyroxene and/or garnet with accessory rutile, ilmenite, apatite, monazite and zircon. They are relatively massive, but there is a weak foliation produced by the arrangement of quartz ribbons. Mafic granulite (orthopyroxene þ clinopyroxene þ plagioclase þ minor quartz) is also a common lithology and is intercalated with other lithologies (Fig. 3f). Garnet gneiss is also an important lithology of the layered gneiss-3. It has the same mineral assemblage as garnet-bearing felsic orthogneiss (i.e. quartz þ mesoperthite þ garnet), but the modal proportion of garnet in the garnet gneiss is much higher than that in the garnet felsic gneiss. Garnet gneiss is commonly accompanied by garnet–sillimanite gneiss. Other rock types, intercalated with the above gneisses, or occurring as thin layers and lenses, are magnetite–quartz gneiss, spinel- and/or sapphirine-bearing siliceous gneiss (impure quartzite), garnet–orthopyroxene gneiss,
GEOLOGY OF MT. RIISER-LARSEN
259
Fig. 3. Photographs of geological structures in the Mt. Riiser-Larsen area. (a) Relatively massive tonalitic orthogneiss layer apparently overlies the felsic orthogneiss dominant layered gneiss-1. (b) Layered gneiss structure is locally cut by Proterozoic mafic dykes and later shear zones. (c) Orthogneisses and paragneisses with layers tens of centimetres to tens of metres thick constitute layered gneiss-3. (d) Proterozoic mafic dykes have intruded and cut the layered gneiss structures, and shear zones are developed along the intrusion plane of mafic dykes. (e) The Riiser-Larsen Main Shear Zone (RLMSZ), composed of mylonite and pseudotachylite, divides the Mt. Riiser-Larsen area into the Main and Western Blocks. (f) Close-up view of the alternation of garnet-bearing felsic gneiss and mafic granulite constituting layered gneiss-2.
sapphirine and/or osumilite-bearing quartzo-feldspathic or aluminous gneiss, and ultramafic rocks including peridotites and pyroxenites. There are two sets of unmetamorphosed mafic dykes, the north –south- and the NE–SW-striking dykes (Figs 2 and 3b, d). The NE –SW dykes, mainly of tholeiitic basalt and high-magnesian andesitic composition, yield Sm–Nd and Rb –Sr isochron ages of 2000–1900 Ma, whereas the
north –south dolerite dykes with alkali basaltic and mid-ocean ridge basalt (MORB)-like tholeiitic composition yield a Rb– Sr isochron age of c.1200 Ma (Suzuki et al. 2008). Mylonitic shear zones, which are rarely accompanied by pseudotachylite, are locally developed, and are commonly of millimetre to centimetre scale. These shear zones are in some cases associated with the zone of mafic dyke intrusions.
260
T. HOKADA ET AL.
A c. 200 m wide pseudotachylite–mylonite zone, the Riiser-Larsen Main Shear Zone (RLMSZ), runs along the west side of Richardson Lake and divides the Mt. Riiser-Larsen area into the Main and Western Blocks (Fig. 2). Its strike is north–south and is subvertical. The gneisses in the RLMSZ are strongly mylonitized and are locally associated with pseudotachylite (Fig. 3e). Kinematic indicators are ambiguous because of the occurrence of more than one phase of movement and it is difficult to estimate displacement of the shear zone in the field or under the microscope. The dominant gneissic structures vary on either side of the RLMSZ. In the Main Block, lithological boundaries generally strike east–west to NE–SW and dip south to SE, whereas in the Western Block, the dip and strike are variable: NE–SW strike with SE dip is dominant in the northern part, whereas NW–SE to NE–SW strike with north dip is distinguished in the southern part of the area. Constituent lithologies are similar on both sides. Similar mylonite–pseudotachylite shear zones are also reported from other areas in the Napier Complex (e.g. Sandiford & Wilson 1984; Sheraton et al. 1987; Motoyoshi 1996; Toyoshima et al. 1999). They are considered to have formed during the retrograde stage, possibly simultaneously with or subsequent to the Proterozoic tectonothermal event that formed the Rayner Complex to the south of the Napier Complex (Sheraton et al. 1987; Harley & Hensen 1990).
Sapphirine 1 quartz-bearing rocks in Mt. Riiser-Larsen Sample description and petrography The association of sapphirine þ quartz is found in a variety of rock types, such as quartzo-feldspathic, siliceous or aluminous gneisses and impure quartzite. The sapphirine þ quartz-bearing gneisses occur as thin layers or lenses in the felsic gneisses or layered gneisses, and in some of them direct contact between sapphirine and quartz is observed. Gneisses at 12 localities (Fig. 2a) contain sapphirine directly in contact with quartz as observed under the microscope, among which the gneisses at five localities (TH97010610, TH97010808, TH97011305, TH97011607, TH97012902–012903; hereafter shortened to 10610, 10808, 11305, 11607, 12902 and 12903, respectively) include thin cordierite, garnet or orthopyroxene–sillimanite coronae occurring between sapphirine and quartz as retrograde products. In other samples (TH97122709, TH97122801, TH97010813, TH97010705, TH97021409, TH970 21326, TH97021329, TH97021330, TH97012810, TH97020713; hereafter shortened to 122709, 122801, 10813, 10705, 21409, 21326, 21329,
21330, 12810, 20713, respectively), no corona textures are observed between sapphirine and quartz. Constituent minerals and textural features of sapphirine þ quartz-bearing gneisses are summarized in Table 3. Aluminous and siliceous gneiss. The sapphirine– quartz association can be found in heterogeneous aluminous or siliceous gneisses composed of quartz, orthopyroxene, garnet, sapphirine, osumilite, sillimanite and feldspar. These gneisses are commonly accompanied by garnet- or garnet–orthopyroxenebearing heterogeneous quartzo-feldspathic gneisses and show a variety of modes of occurrence and mineral assemblages. Direct contact of sapphirine and quartz is observed both in quartz-rich siliceous portions and in sapphirine-rich aluminous portions of the gneisses. The sapphirine–quartz association in some cases forms millimetre- to centimetre-scale clots or seams (Fig. 4a). Sapphirine is generally xenoblastic and less than 1 mm in diameter (Fig. 4b). It occurs interstitially between quartz grains, and is in direct contact with quartz (21326, 21406). In some cases, fine-grained sapphirine forms a symplectitic intergrowth or aggregate with quartz, and this symplectitic texture is interpreted as a pseudomorph after a pre-existing cordierite crystal (Motoyoshi & Hensen 1989) Four samples (11305, 11607, 12902 and 12903) from the Main Block have corona textures in which cordierite and/or garnet formed between sapphirine and quartz crystals (Fig. 5). Quartz is granoblastic with partly rounded form and ranges in size from several hundred micrometres to 1 cm. Usually the width of the corona is constant (c. 100 mm). Xenoblastic or partly granoblastic osumilite, which is commonly replaced by c. 100 mm wide rims of fine-grained (less than 1 mm) symplectite of cordierite–K-feldspar–quartz–orthopyroxene, also occurs interstitially between quartz grains (Fig. 4b). Fine-grained spinel is rarely included in sapphirine, quartz, garnet, osumilite and orthopyroxene in 12903. Sillimanite is locally found in this type of sapphirine–quartz-bearing gneiss, but occurs in a different domain from the sapphirine–quartz association. Biotite occurs locally, but is possibly a retrograde product. Rutile and zircon are minor constituents. Quartzo-feldspathic gneiss. Bluish sapphirine – quartz layers of up to a few centimetres thickness are rarely intercalated within white-coloured quartzo-feldspathic gneiss of up to a few metres to several tens of metres thickness. Common constituents of this type of gneiss are quartz, plagioclase, orthopyroxene and sapphirine, with osumilite and sillimanite also present locally. In these gneisses, compositional layering is developed both in the quartzo-feldspathic and the sapphirine- and/or orthopyroxene-bearing parts (e.g. Hokada et al.
Table 3. Summary of minerals constituting sapphirine– quartz rocks in Mt. Riiser-Larsen Sample no.
Type
Qtz
Pl
M M M M M M M M M M W W W W W W
QF QF S Q Q S A S A S QF QF QF QF QF QF
þ þ þþ þþ þþ þþ þ þþ þ þþ þ þ þ þ þ þ
þ þ
Kfs
L L L L þ þ þ þ þ þ
L
Spr
Opx
þ þ þ L L þ þ þ þ þ þ þ þ þ þ þ
þ þ þ
Os
Spl
Grt
Crd
L L L L
þ L L
þ þ þ þ þ þ þ þ þ þ/R þ/R
þ þ þ þ
L
L p L
Sil
L/R L/R
R R R R
L
Bt L L L L L L
L L L L R R
L þ
Note Spr þ Qtz direct contact Spr þ Qtz direct contact Spr þ Qtz direct contact Spr þ Qtz direct contact Spr þ Qtz direct contact Spr þ Qtz direct contact Spr þ Qtz(þOpx) ! Crd Spr þ Qtz ! Crd Spr þ Qtz þ Opx ! Crd(þGrt) Spr þ Qtz ! Crd, Spr þ Qtz ! Grt Spr þ Qtz direct contact Spr þ Qtz direct contact Spr þ Qtz direct contact Spr þ Qtz direct contact Spr þ Qtz ! Opx þ Sil (Spr þ Qtz ! Opx þ Sil)
GEOLOGY OF MT. RIISER-LARSEN
20713 12820 21326 21329 21330 21406 11305 11607 12902 12903 122709 122801 10813 10705 10610 10808
Area
þ þ, abundant; þ, present; L, local; R, retrograde phase; p, pseudomorph. Area: M, Main Block; W, Western Block. Type (rock type): QF, quartzo-feldspathic; S, siliceous; Q, quartzite; A, aluminous. Mineral abbreviations: Qtz, quartz; Pl, plagioclase; Kfs, alkali feldspar; Spr, sapphirine; Opx, orthopyroxene; Os, osumilite; Spl, spinel; Grt, garnet; Crd, cordierite; Sil, sillimanite; Bt, biotite.
261
262
T. HOKADA ET AL.
Fig. 4. (a) Field appearance of sapphirine–quartz-bearing gneiss in the Mt. Riiser-Larsen area. Sapphirine þ quartz aggregates occur in layers or domains within an orthopyroxene– quartz matrix. (b) Photomicrograph of sapphirine– quartz-rich domain in sapphirine–orthopyroxene –osumilite– quartz gneiss (21326). Sapphirine and quartz are in direct contact with each other, and xenoblastic sapphirine occurs as inclusions in quartz or interstitially between quartz grains. The rim of osumilite is replaced by a fine-grained symplectite of cordierite– K-feldspar–quartz–orthopyroxene. (c) Photomicrograph of sapphirine–quartz-rich layer in sapphirine–orthopyroxene–plagioclase–quartz gneiss (122709). Sapphirine and quartz occur as an aggregate. It is conspicuous that millimetre-scale sapphirine–quartz and plagioclase– quartz(–orthopyroxene) layers are developed. (d) Coexisting orthopyroxene–sillimanite–quartz locally observed in the sapphirine–quartz gneiss (20713). Sillimanite-bearing domains are typically away from the sapphirine–quartz-bearing domain. All photographs are in plane-polarized light. Abbreviations are after Kretz (1983), unless otherwise stated.
2004, fig. 2a). Sapphirine is generally xenoblastic, and is commonly associated with granoblastic quartz (Fig. 4c). The direct contact of sapphirine and quartz is seen in two samples from the Main Block (20713, 12810) and four samples from the Western Block (122709, 122801, 10813, 10705). Sample 10610 from the Western Block has a corona texture of orthopyroxene þ sillimanite between sapphirine and quartz (Fig. 6). Strongly sheared mylonitic quartzo-feldspathic gneiss (10808) in the Western Block also includes a remnant of sapphirine rimmed by an orthopyroxene–sillimanite corona. Orthopyroxene, as a granoblastic grain, is locally found in the sapphirine-bearing layer and is the most common mafic mineral in the quartzo-feldspathic layer with plagioclase and quartz. Osumilite occurs in some of the quartzo-feldspathic gneisses, and is commonly replaced with a 50–100 mm rim
of fine-grained (less than 1 mm) symplectite of cordierite–K-feldspar–quartz–orthopyroxene. Idioblastic sillimanite grains are also found locally in some samples, and are commonly associated with quartz with or without orthopyroxene (Fig. 4d). However, they are not associated with sapphirine þ quartz. Spinel is found as a rare inclusion in sapphirine (10610). Biotite occurs locally, but is possibly a retrograde product. Other minor constituents include rutile and zircon. Impure quartzite. Pale bluish or whitish impure quartzite occurs locally as thin layers or lenses of a few to several tens of centimetres thick, but relatively thick layers of up to 10– 20 m are found in the northwestern ridge of the Mt. Riiser-Larsen area (21329, 21330 in Fig. 2a). Coarse-grained quartz is the dominant constituent, and sapphirine, spinel and sillimanite occur as inclusions or interstitally
GEOLOGY OF MT. RIISER-LARSEN
263
Fig. 5. Mineral textures of the sapphirine–quartz-bearing gneisses in the Main Block of the Mt. Riiser-Larsen area. (a) Photomicrograph of cordierite replacing sapphirine þ quartz (sample 11607). Plane-polarized light. (b) Photomicrograph of sapphirine surrounded by garnet (sample 12903). Plane-polarized light. (c) Backscattered electron image of cordierite replacing sapphirine þ quartz (sample 12902). (d) Backscattered electron image of cordierite and garnet replacing sapphirine þ quartz þ orthopyroxene (sample 12902).
between granoblastic quartz grains. Minor K-feldspar and biotite are also locally seen in the rock. Other sapphirine- and quartz-bearing gneisses. A variety of gneisses with sapphirine and quartz not in direct contact are observed throughout the area,
and are much more common than the abovementioned sapphirine þ quartz-bearing rocks. In such gneisses, sapphirine is commonly surrounded by alkali feldspar, osumilite, orthopyroxene or plagioclase. It also occurs as inclusion in garnet and sillimanite in garnet gneiss or garnet–sillimanite
Fig. 6. Mineral textures of the sapphirine– quartz-bearing gneisses in the Western Block of the Mt. Riiser-Larsen area. (a) Photomicrograph of orthopyroxene þ sillimanite forming between sapphirine and quartz (sample 10610). Planepolarized light. (b) Backscattered electron image of another example of the same reaction texture (sample 10610).
264
Table 4. Representative chemical analyses of constituent minerals in sapphirine– quartz rocks Area: Sample:
M M M M M M M W W M M M TH970TH970- TH970- TH970TH970TH970TH96TH970TH970TH970TH970TH97011305 11607 12902 12903 20713 21326 21406 122709 10610 11607 21326 21406 Point: 133 10 33 132 27 60 92 103 1 5 58 78 Mineral: Spr Spr Spr Spr Spr Spr Spr Spr Spr Os Os Os Spr þ Opx þ Spr þ Opx þ Spr þ Opx Spr þ Opx Spr þ Opx Spr þ Opx þ Spr þ Opx þ Spr þ Opx Spr þ Opx Spr þ Opx þ Spr þ Opx þ Spr þ Opx þ Primary assemblage: Os þ Qtz Os þ Qtz þ Qtz þ Qtz þ Qtz Os þ Qtz Os þ Qtz þ Qtz þ Qtz Os þ Qtz Os þ Qtz Os þ Qtz
13.45 0.00 62.35 0.06 7.77 0.05 0.01 16.69 0.00 0.02 0.00 100.40 O ¼ 20 1.590 0.000 8.685 0.006 0.134 0.634 0.005 0.001 2.941 0.000 0.005 0.000 14.000 0.823 0.175
13.51 0.02 63.66 0.06 6.35 0.09 0.13 17.03 0.00 0.00 0.00 100.85 1.583 0.002 8.792 0.006 0.033 0.589 0.009 0.011 2.975 0.000 0.000 0.000 14.000 0.835 0.053
13.99 0.00 61.85 0.11 7.26 0.00 0.00 16.87 0.00 0.00 0.00 100.08 1.657 0.000 8.634 0.010 0.041 0.678 0.000 0.000 2.979 0.000 0.000 0.000 14.000 0.815 0.057
13.45 0.00 61.27 1.58 7.25 0.03 0.00 16.35 0.04 0.00 0.00 99.97 1.604 0.000 8.610 0.149 0.034 0.689 0.003 0.000 2.906 0.005 0.000 0.000 14.000 0.808 0.047
14.73 0.02 59.59 0.00 8.82 0.12 0.00 16.70 0.00 0.06 0.00 100.04 1.756 0.002 8.370 0.000 0.129 0.750 0.012 0.000 2.967 0.000 0.014 0.000 14.000 0.798 0.147
14.33 0.08 60.86 0.02 7.23 0.00 0.00 17.28 0.00 0.00 0.00 99.80 1.701 0.007 8.514 0.002 0.067 0.650 0.000 0.000 3.058 0.000 0.000 0.000 14.000 0.825 0.094
13.94 0.00 62.27 0.00 7.45 0.10 0.08 16.68 0.00 0.03 0.00 100.55 1.645 0.000 8.661 0.000 0.056 0.679 0.010 0.007 2.935 0.000 0.007 0.000 14.000 0.812 0.076
14.95 0.05 59.73 0.01 8.30 0.11 0.09 16.75 0.01 0.01 0.00 100.01 1.781 0.004 8.388 0.001 0.041 0.786 0.011 0.008 2.976 0.001 0.002 0.000 14.000 0.791 0.050
14.09 0.03 60.35 0.05 9.66 0.07 0.01 15.30 0.04 0.02 0.00 99.62 1.696 0.003 8.561 0.005 0.042 0.930 0.007 0.001 2.746 0.005 0.005 0.000 14.000 0.747 0.043
62.45 0.03 23.40 0.00 1.25 0.10 0.21 8.20 0.01 0.27 4.06 99.98 O ¼ 30 10.256 0.004 4.529 0.000 0.172 0.014 0.025 2.008 0.002 0.086 0.850 17.946 0.921
62.59 0.02 23.68 0.00 1.41 0.04 0.07 8.15 0.00 0.35 4.16 100.47
62.20 0.04 23.65 0.03 1.61 0.00 0.04 8.11 0.00 0.28 4.36 100.32
10.234 0.002 4.563 0.000
10.207 0.005 4.574 0.003
0.193 0.006 0.008 1.987 0.000 0.111 0.867 17.971 0.911
0.221 0.000 0.005 1.983 0.000 0.089 0.913 18.000 0.900
T. HOKADA ET AL.
wt% SiO2 TiO2 Al2O3 Cr2O3 FeO* MnO ZnO MgO CaO Na2O K2O Total Cations Si Ti Al Cr Fe3þ Fe2þ Mn Zn Mg Ca Na K Total XMg XFe3þ
Table 4. Continued Area Sample
M M M M W W W M M M M M TH970TH970TH970TH970TH96TH970TH970TH970TH970TH970TH970TH97011305 11607 12902 21326 122709 10610 10610 12902 12903 11305 12902 12903 Point: 149 23 31 71 106 44 62 12 130 120 23 104 Mineral: Opx Opx Opx Opx Opx Opx Opx Grt Grt Crd Crd Spl Primary Spr þ Opx þ Spr þ Opx þ Spr þ Opx Spr þ Opx Spr þ Opx Spr þ Opx Spr þ Opx Spr þ Opx Spr þ Opx Spr þ Opx þ Spr þ Opx Spr þ Opx þ assemblage: Os þ Qtz Os þ Qtz þ Qtz þ Os þ Qtz þ Qtz þ Qtz þ Qtz þ Qtz þ Os þ Qtz Os þ Qtz þ Qtz Os þ Qtz
49.38 0.08 10.15 0.00 15.20 0.24 0.00 24.75 0.02 0.03 0.00 99.85
Cations
O¼6
Si Ti Al Cr Fe3þ Fe2þ Mn Zn Mg Ca Na K Total XMg XFe3þ
1.775 0.002 0.430 0.000 0.018 0.439 0.007 0.000 1.326 0.001 0.002 0.000 4.000 0.751 0.040
50.39 0.23 10.66 0.00 13.04 0.02 0.09 26.33 0.04 0.00 0.00 100.80
49.07 0.27 10.82 0.00 16.53 0.06 0.08 23.99 0.09 0.02 0.00 100.93
49.35 0.08 11.44 0.00 14.29 0.00 0.13 25.43 0.00 0.03 0.00 100.75
49.74 0.15 10.29 0.00 14.66 0.17 0.00 25.26 0.02 0.02 0.00 100.31
48.26 0.06 10.65 0.02 18.59 0.30 0.00 22.34 0.00 0.00 0.00 100.22
2nd 50.82 0.04 6.38 0.01 18.30 0.17 0.00 24.04 0.00 0.03 0.00 99.78
2nd 40.55 0.01 23.20 0.02 21.36 0.43 0.00 14.55 0.94 0.00 0.00 101.06
2nd 40.89 0.02 23.14 0.01 22.89 0.31 0.00 13.45 0.97 0.01 0.00 101.69
O ¼ 12 1.776 0.006 0.443 0.000 0.000 0.384 0.001 0.002 1.383 0.002 0.000 0.000 3.997 0.783 0.000
1.755 0.007 0.456 0.000 0.021 0.473 0.002 0.002 1.279 0.003 0.001 0.000 4.000 0.730 0.043
1.748 0.002 0.478 0.000 0.024 0.400 0.000 0.003 1.343 0.000 0.002 0.000 4.000 0.771 0.056
1.775 0.004 0.433 0.000 0.011 0.426 0.005 0.000 1.344 0.001 0.001 0.000 4.000 0.759 0.025
1.755 0.002 0.456 0.001 0.029 0.536 0.009 0.000 1.211 0.000 0.000 0.000 4.000 0.693 0.052
1.853 0.001 0.274 0.000 0.020 0.538 0.005 0.000 1.307 0.000 0.002 0.000 4.000 0.708 0.036
2nd 50.66 0.00 34.62 0.00 2.83 0.03 0.03 11.71 0.00 0.00 0.00 99.88
2nd 50.65 0.00 33.84 0.00 3.04 0.02 0.00 11.65 0.01 0.01 0.00 99.22
O ¼ 18
0.04 0.03 59.94 4.10 22.52 0.08 2.27 11.65 0.01 0.07 0.00 100.71 O¼4
2.984 0.001 2.012 0.001
3.006 0.001 2.005 0.001
4.999 0.000 4.027 0.000
5.035 0.000 3.964 0.000
1.314 0.027 0.000 1.596 0.074 0.000 0.000 8.009 0.548
1.407 0.019 0.000 1.474 0.076 0.001 0.000 7.990 0.512
0.234 0.003 0.002 1.723 0.000 0.000 0.001 10.989 0.880
0.253 0.002 0.000 1.727 0.001 0.002 0.000 10.984 0.872
0.001 0.001 1.892 0.087 0.022 0.482 0.002 0.045 0.465 0.000 0.004 0.000 3.000 0.491 0.044
265
*Total Fe as FeO. XMg ¼ Mg/(Mg þ Fe2þ); X(Fe3þ) ¼ Fe3þ/(Fe3þ þ Fe2þ). Mineral abbreviations are as in Table 2. M, Main Block; W, Western Block; 2nd, retrograde product formed from sapphirine þ quartz. Fe3þ for sapphirine, orthopyroxene and spinel is estimated from stoichiometry.
GEOLOGY OF MT. RIISER-LARSEN
wt% SiO2 TiO2 Al2O3 Cr2O3 FeO* MnO ZnO MgO CaO Na2O K2O Total
266
T. HOKADA ET AL.
gneiss. Sapphirine is generally xenoblastic and less than 1 mm in diameter. Quartz is generally granoblastic and ranges in grain size from several hundreds of micrometres to several millimetres. It sometimes occurs as elongated grains up to 1 cm in length. Osumilite surrounding sapphirine is largely replaced by fine-grained (less than 1 mm) symplectites of cordierite –K-feldspar –quartz – orthopyroxene. Sapphirine and quartz, now separated by other minerals, may have stably coexisted at the peak metamorphism.
Reaction textures of cordierite and/or garnet after sapphirine þ quartz in the Main Block Of the 10 sapphirine þ quartz-bearing gneiss samples in the Main Block of Mt. Riiser-Larsen, the corona textures involving cordierite and/or garnet between sapphirine and quartz as retrograde products are observed in four samples (11305, 11607, 12902, 12903). The mode of occurrence of sapphirine and other constituent minerals is similar to that in the samples that show sapphirine –quartz direct contact except for corona textures between sapphirine and quartz, and less commonly between sapphirine and orthopyroxene. The cordierite corona is generally around 50 mm in thickness irrespective of the sample, but locally up to 100 mm (Fig. 5a, c and d). Rarely much coarser-grained (up to 200 – 300 mm) cordierite surrounds sapphirine. The thickness of the garnet corona around sapphirine is similar to that of the cordierite corona, but is more variable from ,10 mm to .300 mm (Fig. 5b and d). In some domains, porphyroblastic garnet has sapphirine inclusions. The variation of the cordierite and/or garnet corona textures around sapphirine crystals in the Main Block may be summarized as follows: (1) a cordierite corona develops between sapphirine and quartz (11305, 11607, 12902, 12903; Fig. 5a); (2) a garnet corona or coarser garnet grain surrounds sapphirine, separating it from quartz (12903; Fig. 5b); (3) a cordierite corona occurs along the grain boundaries between sapphirine and quartz, and sapphirine and orthopyroxene (11305, 12902; Fig. 5c); (4) both cordierite and garnet coronae are developed around sapphirine, and separate it from quartz and orthopyroxene (12902; Fig. 5d).
The width of the orthopyroxene and sillimanite coronae is 10–20 mm, and sillimanite is always formed on the sapphirine side and orthopyroxene on the quartz side. Tiny biotite grains are locally formed in the corona, but it is not clear whether these are synchronous with or later than the formation of the corona texture. An orthopyroxene– sillimanite corona is also seen in the mylonitized sapphirine –quartz-bearing sample 10808. Xenoblastic and fragmented sapphirine grains occur associated with quartz, and a 10 –20 mm orthopyroxene– sillimanite corona similar to that in 10610 is identified in some domains. This sample also has a large proportion of fine-grained biotite grains, but they are commonly restricted to a narrow band within the sample, suggesting that the biotite is a later phase.
Mineral chemistry Chemical analyses of constituent minerals were performed using an electron microprobe with a wavelength-dispersive X-ray analytical system (JEOL JXA-8800M) at the National Institute of Polar Research. Oxide ZAF correction was applied to the analyses. The probe current was kept at about 8 nA with an accelerating voltage at 15 kV. Synthesized pure oxides and natural minerals were used for standards. Representative analyses are listed in Table 4. The chemical features of constituent minerals for the various samples or textural domains are shown in Figures 7 and 8.
Reaction texture of orthopyroxene þ sillimanite after sapphirine þ quartz in the Western Block
Sapphirine. Sapphirine shows a pale blue to colourless pleochroism. Sapphirine compositions are plotted in Figure 7a and b. There are no significant compositional differences between the samples with and without corona textures. The ferric ratio (Fe3þ/total Fe) of most of the sapphirine, calculated from stoichiometry, is less than 0.3 for all samples. Mg/(Mg þ Fe2þ) ratios (XMg) of overall sapphirine analyses range from 0.74 to 0.90. Sapphirine grains surrounded by a cordierite and/or garnet corona have XMg in the range of 0.74 –0.85, and those associated with an orthopyroxene–sillimanite corona have XMg in the range of 0.74– 0.77. Sapphirine compositions of two different corona textures overlap each other. Coarse-grained sapphirine displays weak compositional zoning of increasing Al and decreasing Si toward the rim, similar to that described by Ellis et al. (1980) and Harley (1986).
Two samples from the Western Block show corona textures surrounding sapphirine. An orthopyroxene– sillimanite corona texture is developed between sapphirine and quartz in sample 10610 (Fig. 6).
Orthopyroxene. Overall orthopyroxene has XMg in the range from 0.67 to 0.87. Orthopyroxene in sample 10610, which includes an orthopyroxene– sillimanite corona texture, has lower XMg values,
GEOLOGY OF MT. RIISER-LARSEN (b) 0.45
9.2
Fe3+/total Fe in sapphirine
Al p.f.u. in sapphirine (O = 20)
(a) 9.4
7:9:3
9.0 8.8 8.6 8.4 8.2 8.0 1.3
267
0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05
2:2:1 1.4
1.5
1.6
1.7
1.8
1.9
2.0
0
0.70
0.75
0.80
0.85
0.90
0.95
Mg/(Mg + Fe2+) in sapphirine
Si p.f.u. in sapphirine (O = 20) AlVI p.f.u. in orthopyroxene (O = 6)
(c)
Spr + Qtz in direct contact
0.35
Main Block Spr + Opx + Qtz + Os (21326,21406)
0.30
Spr + Opx + Qtz (20713)
0.25
Western Block Spr + Opx + Qtz (10705,122709,122801)
0.20 0.15
Spr + Qtz with retrograde corona
0.10
Main Block Crd after Spr + Qtz (11305,11607) Crd/Grt after Spr + Qtz (12902,12903)
0.05 0
0.65
0.70
0.75
0.80
0.85
0.90
Western Block Opx + Sil after Spr + Qtz (10610)
Mg/(Mg + Fe2+) in orthopyroxene Fig. 7. (a) Sapphirine compositions in terms of Si and Al cations per 20-oxygen formula unit. (b) Sapphirine compositions in terms of Mg/(Mg þ Fe2þ) ratio (XMg) and Fe3þ/total Fe ratio. (c) Orthopyroxene compositions in terms of XMg and Al cations per six-oxygen formula unit in octahedral site.
in the range of 0.67 –0.72. Orthopyroxene generally shows compositional zoning of decreasing Al2O3 towards the rim. AlVI (¼Si þ Al 2 2 per formula unit (p.f.u.); O ¼ 6) in most of the orthopyroxene analyses is in the range from 0.10 (Al2O3 4.9– 6.1 wt%) to 0.25 (Al2O3 11.3–12.8 wt%), but some of the orthopyroxene core analyses in the aluminous gneiss (12902) have AlVI (p.f.u.) up to 0.28 (Al2O3 12.2–12.7 wt%). In sample 10610, orthopyroxene in the corona texture with sillimanite has considerably lower AlVI (0.07–0.13) than that in the matrix (up to 0.23), and suggests similar but slightly greater XMg in the corona orthopyroxene (Fig. 7c). The ferric ratio (Fe3þ/ total Fe) of the most of the orthopyroxene, calculated from stoichiometry, is less than 0.2 for all samples. Garnet, cordierite, osumilite and spinel. Garnet, which commonly occurs as a corona texture
around sapphirine, in samples 12902 and 12903 is characterized by low grossular (less than 3 mol%) and spessartine (less than 1 mol%) contents, and has XMg in the range from 0.49 to 0.56. Compositional zoning within a single grain is not obvious, but XMg varies depending on the domains in which garnet occurs. Cordierite in the corona texture is characterized by high XMg of 0.86 –0.92. Osumilite shows the highest XMg (0.89 –0.93) among the constituent minerals. No obvious compositional zoning in cordierite and osumilite has been detected. Spinel as inclusions in quartz, orthopyroxene, osumilite and garnet in sample 12903 has variable XMg depending on the mineral enclosing it; it ranges from 0.44 to 0.45 in quartz, from 0.47 to 0.50 in orthopyroxene and osumilite, and from 0.57 to 0.58 in garnet. The ZnO content of spinel is 1.3 wt% in quartz, and 1.7–2.4 wt% in orthopyroxene, osumilite and garnet.
268
T. HOKADA ET AL.
Main Block Spr + Qtz direct contact
Al2O3
Western Block Al2O3
Al2O3
Spr + Qtz direct contact 122709 122801
21326 21406
20713
Spr
Spr
Spr
+Qtz (+Os)
+Qtz
+Qtz
Opx FeO
Crd or Grt after Spr + Qtz
Opx MgO Al2O3
Al2O3
Opx MgO
MgO Al2O3 Opx + Sil after Sil
Spr + Qtz
11305 11607
12902 12903
Spr
10610
Spr Crd
+Qtz (+Os)
+Qtz
Spr Crd +Qtz
Grt Opx FeO
Opx MgO
Opx MgO
MgO
Fig. 8. AFM triangular plots showing the compositional relationships between coexisting minerals. Continuous lines indicate peak assemblages and dashed lines show the retrograde parageneses.
Pressure – temperature constraints by sapphirine – quartz and associated reaction textures Peak and post-peak metamorphic conditions constrained by univariant phase equilibria Phase equilibria in UHT metamorphic conditions, deduced from theoretical thermodynamic calculations (Hensen 1971, 1986, 1987; Ellis et al. 1980; Grew 1982; Holland et al. 1996; Kelsey et al. 2004) and from high-temperature experiments (Hensen & Green 1971, 1972, 1973; Bertrand et al. 1991; Motoyoshi et al. 1993; Audibert et al. 1995; Carrington & Harley 1995a, b), are summarized in Figure 9. The experiments cited above are in reasonable agreement on the topology of
petrogenetic grid and temperatures of the invariant points. However, the pressures deviate systematically between experiments: Carrington & Harley (1995a, b) reported pressures 0.2 GPa lower than those of Hensen & Green (1973), Motoyoshi et al. (1993) and Audibert et al. (1995). This discrepancy is considered to be due to the difficulties of friction correction in the experiments (Carrington & Harley 1995b). Carrington & Harley (1995b) performed cross-calibration experiments using both gas-media and solid-media apparatus. These suggested that the experiments using only solid-media apparatus (Hensen & Green 1973; Motoyoshi et al. 1993; Audibert et al. 1995) had considerably underestimated the required friction correction, thereby introducing a systematic error into their results. For this reason, the pressure conditions of
GEOLOGY OF MT. RIISER-LARSEN
269
Fig. 9. (a) Schematic univariant phase equilibria derived from experimentally constrained reactions, compiled after Hensen & Green (1973), Bertrand et al. (1991), Motoyoshi et al. (1993), Audibert et al. (1995) and Carrington & Harley (1995a, b). Pressure conditions are adjusted based on the discussion by Carrington & Harley (1995a). Stability fields of sapphirine þ quartz and osumilite are shown by vertical stripes and horizontal stripes, respectively. Numbers within rectangles indicate the sequence of transformation of stable parageneses in the Main Block (M) and the Western Block (W) of the Mt. Riiser-Larsen area. (b) The sequence of the stable mineral parageneses in the Western Block. (c) The sequence of the stable mineral parageneses in the Main Block. It should be noted that the stable coexistence of spinel þ quartz in some rocks overlaps with the higher-T part of the sapphirine þ quartz field. Orthopyroxene þ sillimanite þ quartz rarely occurs in the sapphirine–quartz rocks, suggesting that they passed through the higher-pressure side of the [Spl] invariant point.
270
T. HOKADA ET AL.
Carrington & Harley (1995b) are preferable and are used in this study, and the pressure conditions of other experimental results using solid-media apparatus are adjusted to match that of Carrington & Harley (1995b). A variety of mineral parageneses including sapphirine and/or osumilite were used to constrain the peak metamorphic P –T conditions that have been reported for rocks from the Mt. Riiser-Larsen (e.g. Motoyoshi & Matsueda 1984; Makimoto et al. 1989; Motoyoshi & Hensen 1989; Hensen & Motoyoshi 1992; Ishizuka et al. 1998; Hokada 1999; Harley & Motoyoshi 2000; Ishikawa et al. 2000; Suzuki 2000; Suzuki et al. 2001; Hokada et al. 2004; Hokada 2007). Hokada (1999) considered that the stable coexistence of sapphirine þ orthopyroxene þ quartz, sapphirine þ garnet þ quartz and garnet þ osumilite indicates that the peak P–T conditions of the Main Block are restricted in the narrow range of .1050 –1100 8C and 0.6–0.8 GPa (M-1 in Fig. 9). Rare occurrence of spinel (Zn, Cr-free) þ quartz suggests that the peak metamorphic conditions reached or exceeded 1100 8C in a part of the Main Block (M-10 in Fig. 9), whereas in the Western Block, sapphirine þ orthopyroxene þ quartz along with rare occurrence of osumilite is the only observed constraint for peak P–T conditions, giving .1050– 1100 8C and 0.6– 0.9 GPa (W-1 in Fig. 9). However, the absence of garnet þ osumilite and spinel þ quartz, and rare occurrence of osumilite-bearing parageneses in the Western Block might be suggestive of relatively higher-pressure (.0.8– 0.9 GPa) conditions. As described above, different reaction textures after sapphirine þ quartz are observed in the Main and the Western Blocks of the Mt. Riiser-Larsen area. Growth of cordierite and/or garnet between sapphirine and quartz in the Main Block, and of orthopyroxene þ sillimanite in the Western Block indicate the reactions sapphirine þ quartz (þ orthopyroxene) ¼ cordierite
(1)
sapphirine þ quartz (þ orthopyroxene) ¼ garnet
(2)
in the Main Block, and sapphirine þ quartz ¼ orthopyroxene þ sillimanite
(3)
in the Western Block. These contrasting textures imply a marked difference of cooling paths in the Main and Western Blocks. Orthopyroxene þ sillimanite after sapphirine þ quartz in the Western
Block is restricted to higher pressure than 0.8 GPa and temperature drops from .1050 8C (W-1) to 950–1000 8C (W-2 in Fig. 9b). Rare occurrence of osumilite in sapphirine– quartz gneiss suggests that the pressure conditions did not exceed 0.9 GPa, and the reaction texture of osumilite replaced by a fine-grained symplectite of cordierite –K-feldspar – quartz–orthopyroxene indicates further cooling to temperatures below 900–950 8C (W-3 in Fig. 9b). Sapphirine– quartz rocks in the Main Block include a cordierite and/or garnet corona developed along grain boundaries between sapphirine and quartz with or without orthopyroxene, and these reaction textures are consistent with cooling at pressures less than 0.8 GPa (M-1 ! M-2 in Fig. 9c). The rare orthopyroxene þ sillimanite þ quartz association observed locally in the sapphirine–quartz rocks as described above and as reported by Grew (1982) is suggestive of the cooling path in a portion of the Main Block passed through the higher pressure side of the [Spl] invariant point (M-20 in Fig. 9c). The reaction texture of osumilite replaced by a fine-grained symplectite of cordierite–K-feldspar–quartz–orthopyroxene indicates further cooling to temperatures below 900– 950 8C (M-3 in Fig. 9c).
P – T pseudosection calculations for sapphirine – quartz-bearing assemblages P– T pseudosection calculations provide a powerful constraint for the P –T evolution, and have been applied to UHT metamorphic rocks (e.g. Kelsey et al. 2004; Kelsey 2008). In this section, mineral equilibria involving sapphirine –quartzbearing assemblages are discussed with the help of P– T pseudosections. THERMOCALC v.3.26 (tc326) has been used for the calculation of the pseudosections, using the internally consistent thermodynamic dataset (tcds55s.txt, upgrade of Holland & Powell 1998). Pseudosection calculations in this study have been made for the following phases: sapphirine, orthopyroxene, garnet, cordierite, osumilite, sillimanite, quartz, K-feldspar, liquid and water. For simplification, biotite and spinel are excluded from the calculation, because these two phases are rare in sapphirine–quartzbearing rocks. Biotite is rare, and most biotite grains are believed to be secondary phase because they occur locally around orthopyroxene grains or along cracks. Biotite grains are rarely interpreted as primary phase (e.g. Motoyoshi & Hensen 2001) and are characterized by high fluorine contents. Spinel is found in four samples (see Table 3). A tiny spinel grain is observed as an inclusion in sapphirine, but most of the spinel grains occur in different microdomains or layers from the
GEOLOGY OF MT. RIISER-LARSEN
sapphirine–quartz association. Thus, the mineral paragenesis in the sapphirine –quartz-bearing rocks does not reflect the bulk-rock composition but is controlled by the effective local composition. P–T pseudosections calculated using model bulk-rock compositions of sapphirine–orthopyroxene–osumilite– quartz-bearing assemblages are presented in Figure 10. Sapphirine –quartz-bearing rocks are heterogeneous. It is also expected that anatectic melt has been removed from the rocks during metamorphism. To model the sapphirine –quartz-bearing domain of the rocks, the effective bulk-rock compositions have been estimated based on mixtures of sapphirine (15%) þ orthopyroxene (10%) þ osumilite (5%) þ quartz (55%) þ quartzo-feldspathic melt (15%) (Table 5). Five sapphirine–orthopyroxene–osumilite–quartzbearing samples (11305, 11607, 21326, 21406 and 122801) have similar mineral compositions with constant A/AFM ratios (0.43–0.44), except Mg/(Mg þ Fe) ratios (0.76–0.81, average 0.79). A quartzo-feldspathic melt composition (SiO2
0.6
(c)
O px Sp Si rL lC iq rd Q Os tz
[Spr]
900
il z tS Gr pl Qt S r Sp
950
Sp KFMASH r r Grt Sp tz G Op rt x pl Q Cr Liq Opx S d Sp Spr Os Qtz Sil r Li Grt L Sp q Q iq Cr tz [Opx] Spl Crd Qtz dO Os s [Sil]
1000
1050
1100
1150
1200
Grt Spr Opx
H2O = 0.5 wt%, +Qtz + Liq 1.2
(b)
H2O = 1.0 wt%, +Qtz + Liq 1.2 Grt Opx Grt Sil Kfs Sil Kfs
Grt Opx Sil
0.8
Pressure (GPa)
Spr Opx
Opx Sil
Opx Sil Kfs Opx Crd Sil Kfs
850
950
1000
1050
Os 950
Kfs
Grt Spr Opx Kfs
1050
1100
1200
850
Grt Spr Opx
x Os
Opx Sil Crd Os
Opx Crd Kfs
1150
1200
Spr Op
0.8
Os
0.6 1000
1150
Spr Opx Spr Opx Sil Kfs Opx Spr+Qtz Kfs Opx Sil Opx Sil Os Kfs Os Spr Opx Os Kfs
Opx Crd
Temperature (°C)
1100
Grt Opx Sil
1.2
Opx C rd Sil
Opx Sil Crd Os Spr Opx Crd Os Opx Crd Os
900
900
1.3 (d) H2O = 0.1 wt%, +Qtz + Liq
Spr+Qtz
Opx Crd Sil
0.8
850
Opx Crd
0.6
1.0
Spr Opx Crd
0.6
Opx
Opx Crd Kfs
Grt Opx Sil Kfs
1.0
Spr+Qtz
Spr Opx Crd
Opx Crd Sil
Grt Opx Sil Kfs
Opx Crd Kfs
Spr Opx Opx Sil
Opx Sil Kfs
1.0
Grt Spr Opx
Grt Sil
Grt Opx Sil
Opx Crd Sil Kfs
[Crd]
FMAS [Spl] Sp r Opx Qtz G Sp [Sil] rt r Gr t Crd Si Q Gr t Spr l C tz z rd Spl Crd Qt [Opx]
850
KFMASH FMASH FMAS
76.55 wt%, Al2O3 12.95 wt%, K2O 10.5 wt%; the equivalent of qL 30% þ kspL 70%) has been added to the model bulk compositions, because variable amounts of anatectic melt component might have been removed from the system. Another difficulty is the assumption of volatile components (e.g. H2O, CO2, and CH4) in the rocks. In this study, H2O is the only volatile incorporated into the calculation. Three H2O contents (1.0, 0.5 and 0.1 wt%) have been used to assess the effect of the H2O component on the P–T pseudosection. For the high-H2O bulk composition (H2O 1.0 wt%, Fig. 10b), the cordierite stability field is larger (up to .0.95 GPa) and osumilite is not a stable phase, whereas for the medium-H2O bulk composition (H2O 0.5 wt%, Fig. 10c), cordierite is a stable phase at pressures below 0.9 GPa (at .1000 8C) and instead the sapphirine–quartz stability field extends to lower pressure and temperatures. The osumilite stability field appears as a narrow P–T window. For the low-H2O bulk composition (H2O 0.1 wt%,
[Spl]
G Op r t C x S rd O pl s Qt z
0.8
O Gr t px Sil Liq Crd Kfs Qtz Qtz Kfs Sil Opx Grt Os
Opx Crd Kfs Qtz Grt Os Li q
1.0
Opx Sil Kfs Qtz Gr t Os Opx Sil l Gr t Liq Qtz Si tz Crd Os px Q O Gr t r Sp
Opx Crd Kf s Qtz Sil Os Liq
Kelsey et al. (2004) 1.2 THERMOCALC
Qtz s O r Liq Sp il Crd S Gr t
Pressure (GPa)
(a)
271
Spr Opx Crd Os
Spr Opx Crd Opx Crd
Opx Crd Os
900
950
1000
1050
1100
1150
1200
Temperature (°C)
Fig. 10. (a) Univariant phase equilibria calculated by Kelsey et al. (2004) using THERMOCALC with internally consistent thermodynamic dataset. (b –d) P –T pseudosections calculated for the KFMASH model bulk composition for sapphirine– orthopyroxene–osumilite–quartz assemblage with H2O at 1.0, 0.5 and 0.1 wt% using THERMOCALC with internally consistent thermodynamic dataset. Dashed lines indicate the stability limit of sapphirine coexisting with quartz. The osumilite stability field shown by the grey area appears only for relatively H2O-deficient (H2O 0.5 and 0.1 wt%) conditions.
272
T. HOKADA ET AL.
Table 5. Bulk-rock compositions used for P–T pseudosection calculations SiO2
Al2O3
FeO
MgO
K2O
H2O
XMg
A/AFM
FeO þ MgO
Model bulk composition (wt%) calculated from constituent mineral formula Spr 15% þ Opx 10% þ Os 5% þ Qtz 55% þ Liq* 15% 11305 76.57 13.29 2.97 5.30 11607 76.71 13.62 2.30 5.63 21326 76.76 13.43 2.52 5.54 21406 76.71 13.34 2.77 5.44 122801 76.65 13.60 2.03 5.96 Average 76.68 13.45 2.52 5.58
1.79 1.78 1.79 1.79 1.79 1.79
KFMASH model bulk composition (mol%) with varying Spr 15% þ Opx 10% þ Os 5% þ Qtz 55% þ Liq* 15% Average 79.74 8.24 2.19 8.64 H2O ¼ 1.0 wt% 77.07 7.97 2.12 8.35 H2O ¼ 0.5 wt% 78.38 8.10 2.15 8.50 H2O ¼ 0.1 wt% 79.46 8.21 2.18 8.61
0.76 0.81 0.79 0.77 0.83 0.79
0.43 0.44 0.43 0.43 0.43 0.43
0.79 0.79 0.79 0.79
0.43 0.43 0.43 0.43
0.79 0.86 0.64
0.39 0.39 0.39
9.95 9.95 9.95
0.79 0.86 0.64
0.39 0.39 0.39
9.88 9.88 9.88
H2O contents 1.19 1.15 1.17 1.18
FMASH model bulk composition (mol%) with Fe–Mg adjusted Spr 15% þ Opx 10% þ Qtz 75% (anhydrous) Average 83.83 6.23 2.09 7.86 83.83 6.23 1.39 8.55 XMg adjusted to 0.86 XMg adjusted to 0.64 83.83 6.23 3.58 6.37 Spr 15% þ Opx 10% þ Qtz 75% (H2O 0.2 wt%) Average 83.26 6.18 2.07 7.81 XMg adjusted to 0.86 83.26 6.18 1.38 8.50 XMg adjusted to 0.64 83.26 6.18 3.56 6.32
3.35 1.70 0.35
0.68 0.68 0.68
*Liq is SiO2 76.55 wt%, Al2O3 12.95 wt%, K2O 10.5 wt% (qL 30% þ kspL 70%).
Fig. 10d), the sapphirine–quartz stability field extends to lower pressure and temperature (0.75–0.8 GPa, 1000 8C) and cordierite is stable at pressures below 0.8 GPa (at .1000 8C). Osumilite is stable in a wider P–T range (,0.95 GPa, 950–1075 8C) than at higher H2O contents. The above P–T pseudosection calculations include various uncertainties, but can suggest that: (1) the sapphirine –orthopyroxene– quartz assemblage is stable at .0.7 GPa and .1000–1150 8C; (2) the occurrence of osumilite is restricted at relatively low H2O conditions; (3) coexisting sapphirine–orthopyroxene–osumilite–quartz constrains a narrow P– T–X(H2O) range, namely around 0.7–0.9 GPa, 1000–1100 8C and ,0.1 wt% H2O. The differences of sapphirine –quartz breakdown textures can be further analysed by the partial phase diagram calculated for the FMAS(H) system with varying Mg/(Mg þ Fe) ratios (0.86, 0.79 and 0.64) and H2O contents (0 and 0.2 wt%) (Fig. 11 and Table 5). Accordingly, the breakdown of sapphirine þ quartz should occur at 1000– 1100 8C and the product (or products) sapphirine– quartz is (or are) controlled by pressure conditions, bulk Mg/(Mg þ Fe) ratios and H2O contents.
Cordierite is formed after sapphirine þ quartz at lower pressure and higher Mg/(Mg þ Fe), and garnet is restricted to higher pressure and lower Mg/ (Mg þ Fe). Orthopyroxene þ sillimanite is stable between the cordierite and garnet stability fields. In the hydrous system, the cordierite stability field extends toward higher pressure and temperature. Cordierite formed from the decomposition of sapphirine þ quartz is considered as anhydrous in most cases. The replacement of sapphirine –quartz by cordierite, observed in the Main Block, occurs at a lower pressure than orthopyroxene–sillimanite replacing sapphirine–quartz in the Western Block. XMg of cordierite is in the range of 0.8–0.9, and that of garnet is ,0.55. Thus the garnet corona around sapphirine is restricted to a lower Mg/(Mg þ Fe) domain of the rocks, and is consistent with the cordierite corona formation in a higher Mg/(Mg þ Fe) domain. At a given H2O content, the position of the boundary between the cordierite and orthopyroxene þ sillimanite stability fields is almost constant; that is, 0.8– 0.85 GPa at anhydrous condition (Fig. 11a and b) and 1.1–1.2 GPa at H2O 0.2 wt% (Fig. 11d and e).
GEOLOGY OF MT. RIISER-LARSEN
273
(a)
(b)
(c)
1.3
1.3
1.3
1.2
1.2
1.2 Grt
Pressure (GPa)
1.1
Spr + Qtz + Opx
1.0 0.9
W-2
0.6
M-2
M-1
Crd
1000
1100
W-2 M-2
1200
0.8
0.6
anhydrous Mg/(Mg + Fe) = 0.79
1000
M-2
1100
1200
M-1
Crd anhydrous Mg/(Mg + Fe) = 0.64
900
1000
1100
1200
(f) 1.3
Opx + Sil
Opx + Sil Spr + Qtz + Opx
1.1
1.2
Spr + Qtz + Opx
1.1 Crd
1.2 1.1
1.0
1.0
0.9
0.9
0.9
0.8
0.8
0.8
0.7
0.7
0.7
H2O = 0.2 wt% Mg/(Mg + Fe) = 0.86
900 1000 1100 1200 Temperature (°C)
Grt
Crd
1.0
0.6
Spr + Qtz + Opx
0.7
(e) 1.3
1.2
Pressure (GPa)
M-1
Crd
900
Grt
1.0 0.9
W-1
0.8
0.6
(d) 1.3
Spr + Qtz + Opx
Opx + Sil
0.7
anhydrous Mg/(Mg + Fe) = 0.86
900
1.1
1.0 0.9
W-1
0.8 0.7
1.1
Opx + Sil
0.6
H2O = 0.2 wt% Mg/(Mg + Fe) = 0.79
900 1000 1100 1200 Temperature (°C)
0.6
Spr + Qtz + Opx
Crd
H2O = 0.2 wt% Mg/(Mg + Fe) = 0.64
900 1000 1100 1200 Temperature (°C)
Fig. 11. Relative stability fields of Spr þ Opx þ Qtz, Opx þ Sil, Crd and Grt calculated for the FMAS(H) model bulk composition using THERMOCALC with internally consistent thermodynamic dataset. (a– c) Anhydrous FMAS system with XMg ¼ 0.86, 0.79 and 0.64, respectively. (d– f) FMASH system (H2O 0.2 wt%) with XMg ¼ 0.86, 0.79 and 0.64, respectively. It should be noted that the cordierite stability field expands to higher pressure for the hydrous system.
UHT geological structure of Mt. Riiser-Larsen, and implications for UHT metamorphism The geological structure and distribution of the UHT metamorphic rocks (Fig. 2a and c) provide
an insight into the vertical extent (.3–10 km) of the .1000–1100 8C UHT peak metamorphic zone. The metamorphic conditions exceeded 1000 8C for the different structural levels exposed on the Main Block of Mt. Riiser-Larsen. There is a minimum thickness of 4–5 km of UHT-metamorphosed layers, becoming deeper towards the
274
T. HOKADA ET AL.
west in the Main Block. The Western Block represents a c. 0.1–0.3 GPa (c. 3–10 km) deeper structural level than the Main Block. Thus, the metamorphic pressure, and hence the palaeo-depth of metamorphism, is becoming deeper towards the west throughout the area. In addition to the horizontal distribution of this UHT area defined by the stable coexistence of sapphirine –quartz (200 100 km2; Harley & Hensen 1990), our results on the vertical component of the UHT metamorphism provide new constraints for modelling the heat source and tectonic processes of the unusually high-temperature regional metamorphism in the Napier Complex. Maximum metamorphic temperatures estimated for rocks of Mt. Riiser-Larsen are c. 1120–1130 8C (e.g. Harley & Motoyoshi 2000; Ishizuka et al. 2002). The regional occurrence of sapphirine–quartz association reported in this study indicates that metamorphic conditions exceeded at least 1000–1050 8C in .5 km thickness layers of the Mt. Riiser-Larsen area. Two major hypotheses have been put forward regarding the heat source and P–T evolution of the UHT metamorphism in the Napier Complex. Ellis (1987) considered that the quartzo-feldspathic gneiss, which is strongly depleted in heavy rare earth elements, in the Napier Complex requires garnet in the residue from partial melting at 0.8– 1.5 GPa, and that these pressures could be realized only in thickened crust. Based on this, he proposed that the Napier Complex could have formed in an orogenic cycle involving thickened crust, and that the Napier Complex represents the lower part of a doubly thickened crustal segment. Harley (1989, 1991) also suggested that the isobaric cooling obtained for the Napier Complex has resulted from extension of thickened crust. In contrast, Motoyoshi & Hensen (1989) considered that the sapphirine– quartz –orthopyroxene symplectite from Mt. Riiser-Larsen is a pseudomorph after cordierite, and suggested a compressional counterclockwise P–T path for prograde metamorphism, which is contrary to the crustal thickening model proposed by Ellis (1987) and Harley (1989, 1991). Harley & Black (1997) interpreted that the rocks reported by Motoyoshi & Hensen (1989), which had experienced relatively low-pressure conditions, were interleaved in the main body of the Napier Complex and these had together experienced clockwise P –T path during crustal thickening. However, no direct evidence is obtained for the tectonic process and the heat source of the Napier Complex. Intrusion of an anorthosite body, which is occasionally present as thin layers in the complex, has been proposed as the heat source of the UHT metamorphism (Sheraton et al. 1980; Grew 1980), analogous to the common association of anorthosites with charnockites and with osumilite-bearing rocks. In contrast, it has been interpreted that the Napier Complex is the lower plate of doubly thickened
crust, implying a clockwise P–T trajectory (Ellis 1987; Harley 1989, 1991). A similar clockwise P–T path has also been reported from Bunt Island of the Napier Complex (Osanai et al. 2001). Motoyoshi & Hensen (1989) and Hensen & Motoyoshi (1992) discussed polymetamorphic events with a compressional counterclockwise P–T path based on textural evidence, and proposed a regional-scale intrusion of a pluton above the present level of exposure as the heat source of the second event. Harley (1998) noted that convective thinning or detachment of the lithospheric thermal boundary layer during or after crustal thickening may play a major role in causing UHT metamorphism. The proposed tectonic processes to achieve the UHT metamorphism of the Napier Complex are assessed by thermal modelling using numerical solutions (Hokada 1999). Figure 12 shows the results of the calculations. Steady-state geotherms at present and at 2.5 Ga are calculated using the different radioactive heat generation rates between the present and 2.5 Ga (Fig. 12a). A 1D explicit finitedifference method (Peacock 1989) is applied for three models computed by thermal modelling: doubly thickened continental crust produced by continental collision (Fig. 12b); intrusion of hightemperature magma into the shallower level of the Napier metamorphic rocks (Fig. 12c); underplating of asthenospheric mantle by the removal (delamination or detachment) of lithospheric mantle (Fig. 12d). In summary, the process associated with the removal of lithospheric thermal boundary layer and heating directly from hot asthenospheric mantle is one of the possible candidates for the heat source of UHT metamorphism, especially for the regional extent of 1100 8C crustal conditions. Mid- to lower crusts generally contain a certain amount of hydrous minerals, such as biotite and amphibole, and, hence, heating promotes dehydration melting of hydrous minerals. If sufficient heat is supplied, crustal material should generate quartzo-feldspathic magma and relatively mafic restite. If the heat is insufficient, the latent heat of melting compensates and prevents the increase of temperature further above, and the rocks recrystallize in the granulite facies around 700–800 8C, where partial melt and restitic rock could coexist. If already dehydrated dry crust is reheated, partial melting does not occur and the temperature could rise to UHT conditions. Therefore, not only the heat source, but also the dehydration process prior to heating to UHT conditions is considered, to be an essential factor for UHT metamorphism.
Timing of Riiser-Larsen Main Shear Zone As described above, the main UHT event of the Napier Complex is believed to be older than 2480 Ma. Post-dating the UHT event, the thermal
GEOLOGY OF MT. RIISER-LARSEN
(a) 1800
(b) 1800
Steady state geotherm
1400 1200 1000
crust
800
mantle
600 400
2.5 Ga present
200 0
10
20
30
40
50
Doubly thickened crust
1600
Depth of Napier Complex
Temperature (ºC)
Temperature (ºC)
1600
0
275
60
70
1400
80 Ma 60 Ma 50 Ma
1200
30 Ma
800 20 Ma 10 Ma
600 400 200 0
80
0
10
20
30
40
(d) 1800
Magma intrusion
1400
0.1 Ma 1 Ma
0 Ma
1200
10 Ma 30 Ma
1000 800 600 400
1400
0
10
20
30
40
60
70
80
0 Ma
50
60
70
80
Depth (km)
10 Ma 30 Ma 50 Ma
1200
Asthenospheric mantle heating
1000 800 600 400 200
200
1 Ma
0.1 Ma
1600
Temperature (ºC)
1600
50
Depth (km)
(c) 1800
Temperature (ºC)
1 Ma 0.1 Ma
0 Ma
Depth (km)
0
40 Ma
1000
0
mantle
crust 0
10
20
30
40
50
60
70
80
Depth (km)
Fig. 12. One-dimensional finite-difference thermal modelling to simulate the steady-state geotherm (present and 2.5 Ga) and three possible tectonic settings for UHT metamorphism of the Napier Complex modified after Hokada (1999).
regime of the Mt. Riiser-Larsen area was cooling to a steady-state geotherm by 2.38 Ga as evidenced by the Sm –Nd systematics (Suzuki et al. 2006). Subsequently, numerous local rehydration or deformation events are recorded between 2.2 and 0.5 Ga (Suzuki et al. 2001; Hokada & Motoyoshi 2006). Similar early to mid-Proterozoic ages have been reported from other parts of the Napier Complex (Grew et al. 2001; Owada et al. 2001), but it is still not clear whether these events are related to a major tectonothermal event or magmatic event. The Riiser-Larsen Main Shear Zone (RLMSZ) has experienced multiple deformation (at least two stages) with strong mylonite and pseudotachylite fabrics. It is not easy to date the movement of the shear zone directly. Several prominent multiply reactivated shear zones run through the southwestern part of the Mt. Riiser-Larsen area, and the locality of sample 10808 is a branched mylonite– pseudotachylite zone. We have examined zircon and monazite in strongly sheared sapphirine– orthopyroxene–quartz gneiss (TH97010808, hereafter shortened to 10808; Figs 2 and 13). In this sample, an orthopyroxene–sillimanite corona developed around sapphirine grains, consistent
with the other sapphirine –quartz occurrences with an orthopyroxene–sillimanite corona (10610) in the Western Block. We have determined U –Th – Pb ages for monazite and zircon grains from this sheared gneiss sample and for another sapphirinebearing sample (TH97011307; hereafter shortened to 11307) using an electron microprobe at the National Institute of Polar Research, Tokyo, Japan. The analytical protocol and data assessment have been described by Hokada & Motoyoshi (2006). Figure 14 shows the monazite and zircon U–Th– Pb chemical ages of these two samples. Monazite grains in 10808 gave scattered apparent U–Th–Pb ages in the range of 2400–700 Ma, in marked contrast to the sapphirine–orthpopyroxene gneiss (11307), which gave concordant 2500–2450 Ma ages. Monazite ages recorded in sample 10808 gave a wide range of Proterozoic ages and are too scattered to give any clear age constraints. Moreover, monazite grains occur mostly in the biotite-aggregate-rich zone, not in the mylonite–psudotachylite zone. These suggest that the U–Th–Pb disturbance is more strongly associated with the rehydration event predating the mylonite events, and the U–Th–Pb ages may be the oldest limit of the shear zone activity. Therefore, we can tentatively assume that the
276
T. HOKADA ET AL.
Fig. 13. (a) Photomicrograph of mylonitized sapphirine– orthopyroxene– quartz-bearing gneiss (TH97010808). Positions of monazite and zircon are plotted as cross and square, respectively. (b– e) Backscattered electron images of monazite in the sheared gneiss (TH97010808). Numbers indicate U–Th– Pb apparent ages (Ma) analysed by electron microprobe.
GEOLOGY OF MT. RIISER-LARSEN
(a)
277
3500 Mylonitized Spr-Opx-Qtz gneiss 3000
10808
Age (Ma)
2500
2000
1500
1000
1200
Namaqua mnz
1100
500
1000
Monazite
900
Zircon
0 (b)
3500 Spr-Opx gneiss 3000
11307
Age (Ma)
2500
2000
1500
1000
1200
Namaqua monazite
1100 (ref. 1033-1040 Ma)
500
1000
Monazite
900
Zircon
0 Fig. 14. (a) Electron microprobe (EMP) monazite and zircon U–Th–Pb apparent ages of mylonitized sapphirine– orthopyroxene–quartz-bearing gneiss (10808). (b) EMP monazite and zircon apparent ages for sapphirine–orthopyroxene gneiss (11307). Inset: ages of the Namaqua monazites measured in the same analytical session as an internal age-check (reference age 1033–1040 Ma; see details given by Hokada & Motoyoshi 2006).
278
T. HOKADA ET AL.
rehydration and mylonite deformation events recorded in this sample possibly relate to the RLMSZ movement sometime during the Proterozoic.
Implications for the regional metamorphic gradient in the Napier Complex Ellis & Green (1985), Harley (1985) and Sheraton et al. (1987) proposed a near-isobaric cooling path from UHT conditions for rocks of the various localities in the Napier Complex by using garnet – orthopyroxene geothermobarometry. Reaction textures discussed here are consistent with the isobaric cooling path of Harley (1985). The shear zone dividing the area has a north– south strike and subvertical dip. We suggest that the Western Block represents a deeper structural level than the Main Block, and the two blocks were juxtaposed as a result of subvertical movement along the shear zone after isobaric cooling. Unmetamorphosed dolerite dykes that intruded the UHT gneisses at 1.2 Ga were also deformed by the movement of the shear zone. This suggests that juxtaposition of the rocks from different depths occurred after 1.2 Ga.
Using the reaction textures after sapphirine þ quartz as an indicator of the pressure gradient in the Napier Complex, Harley (1998) concluded that sapphirine þ quartz rimmed by coronas involving a combination of cordierite, sillimanite and garnet in the northern areas represent lower pressures at the cooling stage than those of sapphirine þ quartz rimmed by orthopyroxene þ sillimanite in the areas south of Amundsen Bay. Our study shows that both types of reaction textures occur in the Mt. Riiser-Larsen area, and that the rocks of the different structural levels have been juxtaposed as a result of uplift of the Western Block relative to the Main Block along the shear zone after isobaric cooling from UHT conditions. Harley & Hensen (1990) compiled a geo-isobar map, mainly based on garnet– orthopyroxene geobarometry, in which the pressure increases southward in the western part of the Napier Complex. They also related the uplift of the SW part of the Napier Complex to a transpressional tectonic event in the Proterozoic Rayner Complex. Our results and interpretations of the Mt. Riiser-Larsen area support their interpretation but the isobars cannot be as regularly distributed as suggested by the earlier work (Harley & Hensen 1990). Toyoshima et al. (2008) has
Fig. 15. Field pressure gradient of the Napier Complex summarized by Harley & Hensen (1990), and the schematic cross-section proposed in this study, partly modified after Harley & Hensen (1990).
GEOLOGY OF MT. RIISER-LARSEN
suggested that the Napier Complex comprises several small blocks with different strike lines based on the macroscopic geological structures. Our results also suggest that the regional pressure gradient of the Napier Complex is likely to be due to the vertical movement of the blocks along shear zones (Fig. 15). In addition, it should be stressed that because the peak UHT metamorphic conditions (.1000 8C) were apparently attained at different structural levels in the Napier Complex, a large heat source is necessary to give rise to UHT conditions on this observed scale. Our thanks go to the following people: M. Satish-Kumar for his thoughtful and insightful editorial comments, and E. S. Grew and D. E. Kelsey for their critical reviews, which improved the manuscript considerably. Comments were also made by N. M. Kelly, K. Shiraishi, Y. Hiroi and Y. Osanai, which helped in preparing and revising the manuscript. P –T pseudosection calculation using THERMOCALC was kindly taught by C. Clarke. An earlier draft was reviewed and improved by B. J. Hensen and A. P. Nutman. The samples used in this study were collected during the geological fieldwork of the 1996– 1997 Japanese Antarctic Research Expedition (JARE), and the members of JARE are thanked for their support for our field operation.
References A SAMI , M., S UZUKI , K., G REW , E. S. & A DACHI , M. 1998. CHIME ages for granulites from the Napier Complex, East Antarctica. Polar Geoscience, 11, 172–199. A SAMI , M., S UZUKI , K. & G REW , E. S. 2002. Chemical Th– U– total Pb dating by electron microprobe analysis of monazite, xenotime and zircon from the Archaean Napier Complex, East Antarctica: evidence for ultra-high-temperature metamorphism at 2400 Ma. Precambrian Research, 114, 249–275. A UDIBERT , N., H ENSEN , B. J. & B ERTRAND , P. 1995. Experimental study of phase relations involving osumilite in the system K2O– FeO–MgO– Al2O3 – SiO2 –H2O at high pressure and temperature. Journal of Metamorphic Geology, 13, 331 –344. B ERTRAND , P., E LLIS , D. J. & G REEN , D. H. 1991. The stability of sapphirine– quartz and hypersthene–sillimanite–quartz assemblages: an experimental investigation in the system FeO–MgO– Al2O3 – SiO2 under H2O and CO2 conditions. Contributions to Mineralogy and Petrology, 108, 55– 71. B LACK , L. P., W ILLIAMS , I. S. & C OMPSTON , W. 1986. Four zircon ages from one rock: the history of a 3930 Ma-old granulite from Mount Sones, Antarctica. Contributions to Mineralogy and Petrology, 94, 427– 437. B OWRING , S. A., W ILLIAMS , I. S. & C OMPSTON , W. 1989. 3.96 Ga gneisses from the Slave province, Northwest Territories, Canada. Geology, 17, 971– 975. C ARRINGTON , D. P. & H ARLEY , S. L. 1995a. Partial melting and phase relations in high-grade metapelites: an experimental petrogenetic grid in the KFMASH
279
system. Contributions to Mineralogy and Petrology, 120, 270–291. C ARRINGTON , D. P. & H ARLEY , S. L. 1995b. The stability of osumilite in metapelitic granulites. Journal of Metamorphic Geology, 13, 613–625. D ALLWITZ , W. B. 1968. Coexisting sapphirine and quartz in granulites from Enderby Land, Antarctica. Nature, 219, 476–477. E LLIS , D. J. 1987. Origin and evolution of granulites in normal and thickened crusts. Geology, 15, 167– 170. E LLIS , D. J. & G REEN , D. H. 1985. Garnet-forming reactions in mafic granulites from Enderby Land, Antarctica—implications for geothermometry and geobarometry. Journal of Petrology, 26, 633–662. E LLIS , D. J., S HERATON , J. W., E NGLAND , R. N. & D ALLWITZ , W. B. 1980. Osumilite –sapphirine– quartz granulites from Enderby Land, Antarctica: mineral assemblages and reactions. Contributions to Mineralogy and Petrology, 72, 123– 143. F ITZSIMONS , I. C. W. 2000. A review of tectonic events in the East Antarctic Shield and their implications for Gondwana and earlier supercontinents. Journal of African Earth Sciences, 31, 3 –23. G REW , E. S. 1980. Sapphirine þ quartz association from Archaean rocks in Enderby Land, Antarctica. American Mineralogist, 65, 821– 836. G REW , E. S. 1982. Osumilite in the sapphirine –quartz terrane of Enderby Land, Antarctica: implications for osumilite petrogenesis in the granulite facies. American Mineralogist, 67, 762– 787. G REW , E. S. 1998. Boron and beryllium minerals in granulite-facies pegmatites and implications of beryllium pegmatites for the origin and evolution of the Archean Napier Complex of East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 53, 74–93. G REW , E. S., S UZUKI , K. & A SAMI , M. 2001. CHIME ages of xenotime, monazite and zircon from beryllium pegmatites in the Napier Complex, Khmara Bay, Enderby Land, East Antarctica. Polar Geoscience, 14, 99–118. H ARLEY , S. L. 1983. Regional geobarometry– geothermometry and metamorphic evolution of Enderby Land, Antarctica. In: O LIVER , R. L., J AMES , P. R. & J AGO , J. B. (eds) Antarctic Earth Science. Australian Academy of Science, Canberra, 25– 30. H ARLEY , S. L. 1985. Garnet–orthopyroxene bearing granulites from Enderby Land, Antarctica: metamorphic pressure–temperature–time evolution of the Archaean Napier Complex. Journal of Petrology, 26, 819–856. H ARLEY , S. L. 1986. A sapphirine– cordierite–garnet– sillimanite granulite from Enderby Land, Antarctica: implications for FMAS petrogenetic grids in the granulite facies. Contributions to Mineralogy and Petrology, 94, 452–460. H ARLEY , S. L. 1987. A pyroxene-bearing metaironstone and other pyroxene-granulites from Tonagh Island, Enderby Land, Antarctica: further evidence for very high temperature (.980 8C) Archaean regional metamorphism in the Napier Complex. Journal of Metamorphic Geology, 5, 341 –356. H ARLEY , S. L. 1989. The origins of granulites: a metamorphic perspective. Geological Magazine, 126, 215– 247.
280
T. HOKADA ET AL.
H ARLEY , S. L. 1991. The crustal evolution of some East Antarctic granulites. In: T HOMSON , M. R. A., C RAME , J. A. & T HOMSON , J. W. (eds) Geological Evolution of Antarctica. Cambridge University Press, Cambridge, 7– 12. H ARLEY , S. L. 1998. On the occurrence and characterization of ultrahigh-temperature crustal metamorphism. In: T RELOAR , P. J. & O’B RIEN , P. J. (eds) What Drives Metamorphism and Metamorphic Reactions? Geological Society, London, Special Publications, 138, 81–107. H ARLEY , S. L. 2003. Archaean to Pan-African crustal development and assembly of East Antarctica: metamorphic characteristics and tectonic implications. In: Y OSHIDA , M., W INDLEY , B. F. & D ASGUPTA , S. (eds) Proterozoic East Gondwana: Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 203– 230. H ARLEY , S. L. 2004. Extending our understanding of ultrahigh temperature crustal metamorphism. Journal of Mineralogical and Petrological Sciences, 99, 140– 158. H ARLEY , S. L. & B LACK , L. P. 1997. A revised Archaean chronology for the Napier Complex, Enderby Land, from SHRIMP ion-microprobe studies. Antarctic Science, 9, 74–91. H ARLEY , S. L. & H ENSEN , B. J. 1990. Archaean and Proterozoic high-grade terranes of East Antarctica (40–808E): a case study of diversity in granulite facies metamorphism. In: A SHWORTH , J. R. & B ROWN , M. (eds) High-temperature Metamorphism and Crustal Anatexis. Unwin Hyman, London, 320–370. H ARLEY , S. L. & M OTOYOSHI , Y. 2000. Al zoning in orthopyroxene in a sapphirine quartzite: evidence for .1120 8C UHT metamorphism in the Napier Complex, Antarctica, and implications for the entropy of sapphirine. Contributions to Mineralogy and Petrology, 138, 293 –307. H ARLEY , S. L., K INNY , P. D., S NAPE , I. & B LACK , L. P. 2001. Zircon chemistry and the definition of events in Archaean granulite terrains. In: Extended Abstracts of 4th International Archaean Symposium. AGSOGeoscience Australia Record 2001/37, 511– 513. H ENSEN , B. J. 1971. Theoretical phase relations involving cordierite and garnet in the system MgO –FeO– Al2O3 –SiO2. Contributions to Mineralogy and Petrology, 33, 191–214. H ENSEN , B. J. 1986. Theoretical phase relations involving cordierite and garnet revisited: the influence of oxygen fugacity on the stability of sapphirine and spinel in the system Mg–Fe–Al– Si–O. Contributions to Mineralogy and Petrology, 92, 362– 367. H ENSEN , B. J. 1987. P –T grids for silica-undersaturated granulites in the system MAS (n þ 4) and FMAS (n þ 3)—tools for the derivation of P –T paths of metamorphism. Journal of Metamorphic Geology, 5, 255– 271. H ENSEN , B. J. & G REEN , D. H. 1971. Experimental study of the stability of cordierite and garnet in pelitic compositions at high pressures and temperatures. I. Compositions with excess alumino-silicate. Contributions to Mineralogy and Petrology, 33, 309–330. H ENSEN , B. J. & G REEN , D. H. 1972. Experimental study of the stability of cordierite and garnet in pelitic
compositions at high pressures and temperatures. II. Compositions without excess alumino-silicate. Contributions to Mineralogy and Petrology, 35, 331–354. H ENSEN , B. J. & G REEN , D. H. 1973. Experimental study of the stability of cordierite and garnet in pelitic compositions at high pressures and temperatures. III. Synthesis of experimental data and geological applications. Contributions to Mineralogy and Petrology, 38, 151–166. H ENSEN , B. J. & M OTOYOSHI , Y.1992. Osumiliteproducing reactions in high-temperature granulites from the Napier Complex, East Antarctica: tectonic implications. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Science. Terra, Toyko, 87–92. H OKADA , T. 1999. Thermal evolution of the UHT metamorphic rocks in the Napier Complex, East Antarctica. PhD thesis, The Graduate University for Advanced Studies, Tokyo. H OKADA , T. 2001. Feldspar thermometry in ultrahigh-temperature metamorphic rocks: evidence of crustal metamorphism attaining 1100 8C in the Archean Napier Complex, East Antarctica. American Mineralogist, 86, 932–938. H OKADA , T. 2007. Perrierite in sapphirine– quartz gneiss: geochemical and geochronological features and implications for accessory-phase paragenesis of UHT metamorphism. Journal of Mineralogical and Petrological Sciences, 102, 44– 49. H OKADA , T. & M OTOYOSHI , Y. 2006. Electron microprobe technique for U– Th–Pb and REE chemistry of monazite, and its implications for pre-, peak- and postmetamorphic events of the Lu¨tzow-Holm Complex and the Napier Complex, East Antarctica. Polar Geoscience, 19, 118– 151. H OKADA , T. & S UZUKI , S. 2006. Feldspar in felsic orthogneiss as indicator for UHT crustal processes. Journal of Mineralogical and Petrological Sciences, 101, 260– 264. H OKADA , T., M ISAWA , K., S HIRAISHI , K. & S UZUKI , S. 2003. Mid to Late Archaean (3.3–2.5 Ga) tonalitic crustal formation and high-grade metamorphism at Mt. Riiser-Larsen, Napier Complex, East Antarctica. Precambrian Research, 127, 215–228. H OKADA , T., M ISAWA , K., Y OKOYAMA , K., S HIRAISHI , K. & Y AMAGUCHI , A. 2004. SHRIMP and electron microprobe chronology of UHT metamorphism in the Napier Complex, East Antarctica: implications for zircon growth at .1000 8C. Contributions to Mineralogy and Petrology, 147, 1 –20. H OLLAND , T. J. B. & P OWELL , R. 1998. An internally consistent thermodynamic data set for phases of petrological interest. Journal of Metamorphic Geology, 16, 309– 343. H OLLAND , T. J. B., B ABU , E. V. S. S. K. & W ATERS , D. J. 1996. Phase relations of osumilite and dehydration melting in pelitic rocks: a simple thermodynamic model for the KFMASH system. Contributions to Mineralogy and Petrology, 124, 383–394. I SHIKAWA , M., H OKADA , T., I SHIZUKA , H., M IURA , H., S UZUKI , S. & T AKADA , M. 2000. Geological map of Mount Riiser-Larsen, Enderby Land, Antarctica.
GEOLOGY OF MT. RIISER-LARSEN Antarctic Geological Map Series Sheet 37. National Institute of Polar Research, Tokyo. I SHIZUKA , H. 2008. An overview of geological studies of JARE in the Napier Complex, Enderby Land, East Antarctica. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: a Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 121–137. I SHIZUKA , H., I SHIKAWA , M., H OKADA , T. & S UZUKI , S. 1998. Geology of the Mt. Riiser-Larsen area of the Napier Complex, Enderby Land, East Antarctica. Polar Geoscience, 11, 154– 171. I SHIZUKA , H., S UZUKI , S. & N AKAMURA , A. 2002. Peak temperatures of ultra-high temperature metamorphism of the Napier Complex, Enderby Land, East Antarctica, as deduced from porphyroclastic pyroxenes of meta-ultramafic rocks. Polar Geoscience, 15, 1– 16. K ELLY , N. M. & H ARLEY , S. L. 2005. An integrated microtextural and chemical approach to zircon geochronology: refining the Archaean history of the Napier Complex, east Antarctica. Contributions to Mineralogy and Petrology, 149, 57–84. K ELLY , N. M., C LARKE , G. L. & F ANNING , C. M. 2005. Archaean crust in the Rayner Complex of East Antarctica: Oygarden Group of Islands, Kemp Land. Transactions of the Royal Society of Edinburgh, Earth Science, 95, 491–510. K ELSEY , D. E. 2008. On ultrahigh-temperature crustal metamorphism. Gondwana Research, 19, 152 –172. K ELSEY , D. E., W HITE , R. W., H OLLAND , T. J. B. & P OWELL , R. 2004. Calculated phase equilibria in K2O –FeO–MgO –Al2O3 –SiO2 – H2O for sapphirine–quartz-bearing mineral assemblages. Journal of Metamorphic Geology, 22, 559– 578. K RETZ , R. 1983. Symbols for rock-forming minerals. American Mineralogist, 68, 277–279. M AKIMOTO , H., A SAMI , M. & G REW , E. S. 1989. Some geological observations on the Archean Napier Complex at Mt. Riiser-Larsen, Amundsen Bay, Enderby Land. Proceedings of NIPR Symposium on Antarctic Geosciences, 3, 128– 141. M IYAMOTO , T., Y OSHIMURA , Y., D UNKLEY , D. J., M OTOYOSHI , Y. & C ARSON , C. J. 2006. Mineral ages for multi-isotope system in phlogopite-bearing pyroxene granulite and felsic gneiss, the Howard Hills, Enderby Land, East Antarctica: Possible Proterozoic tectonothermal events in the Napier Complex. Polar Geoscience, 19, 152– 172. M OTOYOSHI , Y. 1996. Pseudotachylite from McIntyre Island, Enderby Land, East Antarctica: evidence for a rapid crystallization. Proceedings of NIPR Symposium on Antarctic Geoscience, 9, 65–75. M OTOYOSHI , Y. & H ENSEN , B. J. 1989. Sapphirine– quartz–orthopyroxene symplectites after cordierite in the Archaean Napier Complex, Antarctica: evidence for a counterclockwise P –T path? European Journal of Mineralogy, 1, 467– 471. M OTOYOSHI , Y. & H ENSEN , B. J. 2001. F-rich phlogopite stability in ultra-high-temperature metapelites from the Napier Complex, East Antarctica. American Mineralogist, 86, 1404–1413.
281
M OTOYOSHI , Y. & M ATSUEDA , H. 1984. Archean granulites from Mt. Riiser-Larsen in Enderby Land, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 33, 103–125. M OTOYOSHI , Y., H ENSEN , B. J. & A RIMA , M. 1993. Experimental study of the high-pressure stability limit of osumilite in the system K2O–MgO –Al2O3 – SiO2: implications for high-temperature granulites. European Journal of Mineralogy, 5, 439–445. O SANAI , Y., T OYOSHIMA , T., O WADA , M., T SUNOGAE , T., H OKADA , T., C ROWE , W. A. & K USACHI , I. 2001. Ultrahigh temperature sapphirine–osumilite and sapphirine–quartz granulites from Bunt Island in the Napier Complex, East Antarctica—Reconnaissance estimation of P–T evolution. Polar Geoscience, 14, 1–24. O WADA , M., O SANAI , Y., T SUNOGAE , T., H AMAMOTO , T., K AGAMI , H., T OYOSHIMA , T. & H OKADA , T. 2001. Sm–Nd garnet ages of retrograde garnet bearing granulites from Tonagh Island in the Napier Complex, East Antarctica: a preliminary study. Polar Geoscience, 14, 75– 87. P EACOCK , S. M. 1989. Thermal modeling of metamorphic pressure–temperature–time paths: a forward approach. In: S PEAR , F. S. & P EACOCK , S. M. (eds) Metamorphic Pressure–Temperature–Time Paths. American Geophysical Union, Short Courses in Geology, 57–102. S ANDIFORD , M. & P OWELL , R. 1986. Pyroxene exsolution in granulites from Fyfe Hills, Enderby Land, Antarctica: evidence for 1000 8C metamorphic temperatures in Archaean continental crust. American Mineralogist, 71, 946–954. S ANDIFORD , M. & P OWELL , R. 1988. Pyroxene exsolution in granulites from Fyfe Hills, Enderby Land, Antarctica: evidence for 1000 8C metamorphic temperatures in Archaean continental crust—Reply. American Mineralogist, 73, 434– 438. S ANDIFORD , M. & W ILSON , C. J. L. 1984. The structural evolution of the Fyfe Hills – Khmara Bay region, Enderby Land, East Antarctica. Australian Journal of Earth Sciences, 31, 403 –426. S HERATON , J. W., T INGEY , R. J., B LACK , L. P., O FFE , L. A. & E LLIS , D. J. 1987. Geology of Enderby Land and western Kemp Land, Antarctica. Bulletin, Australian Bureau of Mineral Resources, 223, 1– 51. S UZUKI , S. 2000. Geochemistry and geochronology of ultra-high temperature metamorphic rocks from the Mt. Riiser-Larsen area in the Archaean Napier Complex, East Antarctica. PhD thesis, The Graduate University for Advanced Studies, Tokyo. S UZUKI , S., K AGAMI , H., I SHIZUKA , H. & H OKADA , T. 2001. Sm–Nd mineral isochron ages of sapphirine–quartz gneiss from the Mt. Riiser-Larsen area in the Napier Complex, East Antarctica. Polar Geoscience, 14, 88–98. S UZUKI , S., A RIMA , M., W ILLIAMS , I. S., S HIRAISHI , K. & K AGAMI , H. 2006. Thermal history of UHT metamorphism in the Napier Complex, East Antarctica: insights from zircon, monazite, and garnet ages. Journal of Geology, 114, 65– 84. S UZUKI , S., I SHIZUKA , H. & K AGAMI , H. 2008. Early to middle Proterozoic dykes in the Mt. Riiser-Larsen area of the Napier Complex, East Antarctica: tectonic implications as deduced from geochemical studies. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds)
282
T. HOKADA ET AL.
Geodynamic Evolution of East Antarctica: a Key to the East–West Gondwana Connection. Geological Society, London, Special Publications, 308, 195–210. T OYOSHIMA , T., O SANAI , Y., O WADA , M., T SUNOGAE , T., H OKADA , T. & C ROWE , W. A. 1999. Deformation of ultrahigh-temperature metamorphic rocks from Tonagh Island in the Napier Complex, East Antarctica. Polar Geoscience, 12, 29–48.
T OYOSHIMA , T., O SANAI , Y. & N OGI , Y. 2008. Macroscopic geological structures of the Napier and Rayner Complexes, East Antarctica. In: S ATISH K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: a Key to the East–West Gondwana Connection. Geological Society, London, Special Publications, 308, 139– 146.
Early Palaeozoic metasomatism of the Archaean Napier Complex, East Antarctica CHRISTOPHER J. CARSON1,2 & JAY J. AGUE1 1
Department of Geology and Geophysics, Yale University, New Haven, CT 06511, USA
2
Present address: Geoscience Australia, GPO Box 378, Canberra, A.C.T. 2601, Australia (e-mail:
[email protected]) Abstract: Emplacement of post-tectonic Early Palaeozoic pegmatites on Tonagh Island, Napier Complex, East Antarctica, was accompanied by the introduction of aqueous low-salinity fluids at mid-P upper-amphibolite facies conditions (c. 8 kbar, c. 680 8C). Fluid–wall-rock interaction resulted in the development of spectacular alteration selvedges, in the immediate vicinity of the pegmatites, in adjacent Archaean orthogneisses. Archaean wall-rocks affected by the infiltration of aqueous fluids show contrasting patterns of K, Na and Ca metasomatism, which we demonstrate was fundamentally controlled by disequilibrium of invading fluids with wall-rock feldspars, rather than fluid flow up or down regional pressure or temperature gradients. Other species, such as the rare earth elements (REE; except Eu), Y, P, Rb, Th, U and Pb, for example, show significant enrichment in the metasomatized wall-rock in both examples studied. Enrichment of P (and Y), sympathetic with that of REE enrichment, is consistent with recent suggestions that REE transport in fluids may be enhanced by complexing with dissolved P and Y compounds. Dehydration and partial melting of previously unmetamorphosed sedimentary rocks, underthrust beneath the SW Napier Complex, has been long considered a viable source for felsic pegmatites (and associated fluids) observed in that region. Our results are consistent with the hypothesis that underplated sedimentary rocks were a viable source for pegmatite melt and aqueous fluids. Furthermore, as our study demonstrates a plausible relationship between Early Palaeozoic (c. 500–530 Ma) pegmatites and fluid infiltration, we suggest that dehydration and prograde partial melting of the underthrust sedimentary rocks beneath the Napier Complex occurred, at least in part, by convergent Early Palaeozoic tectonism.
Fluids are an integral component of many crustal metamorphic environments, participating in and promoting a wide variety of metamorphic reactions and mechanical processes. Fluid-saturated conditions greatly enhance rates of grain boundary diffusion, facilitating recrystallization processes (e.g. Ayers et al. 1999) and the development of new mineral assemblages (e.g. Jamtveit et al. 1990; Klaper 1990; Rubie 1990; Rockow et al. 1997; Carson et al. 2000) and foliations (e.g. Beach 1976; White et al. 1980; Passchier 1985). In particular, studies have highlighted the impact of fluid flow and open-system behaviour on the evolution of crustal rocks on a regional scale (e.g. Ferry & Dipple 1991; Dipple & Ferry 1992; Ague 1994a, b; Ferry 1996; Ferry & Gerdes 1998; Ague & Rye 1999; Breeding & Ague 2002). These studies recognized that the generation and passage of large volumes of hydrous fluids during the evolution of metamorphic terranes are common and can act as a primary vector for transfer of heat (e.g. Bickle & McKenzie 1987; Chamberlain & Rumble 1988) and of non-volatile elements (e.g. Beach 1980; Ferry & Dipple 1991; Manning 1994) resulting in significant modification of bulk-rock compositions (e.g. Ague 1991, 1997,
1998; Ague & van Haren 1996), even on a crustal scale (Breeding & Ague 2002). Alkali and alkali earth metasomatism is a common feature of many terranes affected by fluid influx. As the thermodynamic behaviour of these species in aqueous systems is relatively well known for typical crustal conditions (e.g. Sverjensky et al. 1991), the alkali and alkali earth geochemistry of metasomatized rocks has been used as an important indicator for fluid flow direction relative to thermal and baric gradients in metasomatic regimes (e.g. Orville 1962; Dipple & Ferry 1992; Ague 1997). In this paper we present a mineralogical, petrological, geochemical and geochronological analysis of two examples of upper amphibolitefacies metasomatism adjacent to Early Palaeozoic post-tectonic peraluminous pegmatites within the Archaean Napier Complex of East Antarctica. We examine the extent of non-volatile mass transfer during fluid –rock interaction with emphasis on the behaviour of K, Na and Ca, quantify mass and volume changes, and discuss the implications of fluid influx into the Napier Complex for regional tectonic models. We also highlight the contrasting
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 283 –316. DOI: 10.1144/SP308.14 0305-8719/08/$15.00 # The Geological Society of London 2008.
284
C. J. CARSON & J. J. AGUE
K, Na and Ca behaviour during metasomatism in the two examples studied and emphasize the fundamental importance of precursor feldspar mineralogy in controlling patterns of K, Na and Ca metasomatism.
Geological setting The Archaean Napier Complex, East Antarctica, is located between the latitudes 67.58 and 66.08S and longitudes 488 and 538E (Fig. 1). This high-grade regional metamorphic terrane has been the focus of considerable geological interest over the last four decades, primarily because of the recognition of extreme peak regional metamorphic conditions exceeding 1100 8C (e.g. Dallwitz 1968; Ravich 1972; Ellis 1980; Ellis et al. 1980; Grew 1980, 1982; Sheraton et al. 1980, 1987; Sandiford & Powell 1986; Harley 1987, 2003; Harley & Hensen 1990; Harley & Motoyoshi 2000; Hokada 2001) synchronous with the development of an intense gneissosity and isoclinal folding (D1 –D2). As a consequence, the majority of the geological and chronological investigations of the Napier Complex have generally focused on characterization of ultrahigh-temperature (UHT) tectonothermal events and processes. In contrast, post-UHT events within the Napier Complex have received comparatively little attention and, as such, the
post-UHT metamorphic, structural and geochronological evolution is not well understood. It is widely acknowledged that the Napier Complex underwent upper amphibolite-facies metamorphism postdating peak UHT metamorphism. Development of amphibolite-facies assemblages has been associated with upright open folding (D3) at c. 2470 Ma, long after UHT metamorphism (e.g. Black & James 1983; Black et al. 1983a; Harley & Black 1987; Black 1988). Other workers, however, have argued that UHT metamorphism occurred at c. 2470 Ma, with upright D3 folding evolving during the waning stages of peak metamorphism (e.g. DePaolo et al. 1982; Grew 1998; Carson et al. 2002a). Amphibolite-facies conditions also prevailed during the development of narrow shear zones that postdate D3 upright folding (D4 –D5; e.g. Harley 1985; Sandiford 1985; Harley & Hensen 1990), resulting in recrystallization and destruction of UHT mineral assemblages (e.g. Sheraton et al. 1987; Ellis & Green 1985) and the formation of hydrous mineral assemblages (e.g. Harley 1985; Tsunogae et al. 1999). The absolute timing of the development of shear zones and upper amphibolitefacies metamorphism has remained somewhat uncertain. A common conclusion is that such features developed coevally with pervasive high-grade tectonism within the adjacent Neoproterozoic (c. 1000 Ma) Rayner Complex, which bounds the Napier Complex to the south and southeast
Fig. 1. Location of Tonagh Island (indicated in box) within the Archaean Napier Complex. Boundary between Napier and Neoproterozoic Rayner Complexes (dashed with question marks) is after Harley & Black (1987) and Harley & Hensen (1990). The cross-section discussed in the text and Figure 10 is indicated as A –A0 . Upper inset shows location of the Napier Complex within the Antarctic continent.
METASOMATISM OF THE NAPIER COMPLEX
(Fig. 1). This interpretation is supported by U– Pb zircon and monazite data from the SW Napier Complex that show isotopic disturbance of Rayner age (e.g. Black et al. 1983a, b, 1984, 1986b; Grew et al. 1982), which probably reflects the thermal effects of c. 1000 Ma tectonic activity from the adjacent Rayner Complex. For example, Harley (1985) determined a c. 1000 Ma age for the development of upper amphibolite-facies shear zones in the Casey Bay region. In contrast, Sandiford (1985) proposed two episodes of postArchaean shear zone development between c. 1000 and c. 500 Ma. The latter event described by Sandiford (1985) involved reactivation of shear zones and was accompanied by emplacement of significant volumes of pegmatite at amphibolite-facies grade. Furthermore, Sandiford (1985) has, to date, provided the only description of geochemical modification of Archaean protolith as a result of fluid influx associated with pegmatite emplacement and shear zone development. He also suggested that such pegmatites provide a major mechanism by which fluid may have been reintroduced into the terrane after effective dehydration by UHT metamorphism. Although no specific geochronological
285
data for the features described by Harley (1985) or Sandiford (1985) were presented, it is probable that the latter, pegmatite-related event described by Sandiford (1985) represents c. 500 Ma felsic igneous activity that is common in the Casey Bay region (e.g. Black et al. 1983a). Several workers have reported the emplacement of Early Palaeozoic planar felsic pegmatites and mafic lamproite dykes into the Napier Complex (Grew 1978; Grew & Manton 1979; Black et al. 1983a; Sheraton et al. 1987; Miyamoto et al. 2000; Carson et al. 2002b). Grew (1978) and Grew & Manton (1979) noted the development of amphibolite-facies ‘alteration’ zones or ‘bleach’ zones affecting host lithologies adjacent to felsic pegmatites of presumed Early Palaeozoic age. These amphibolite-facies metasomatic ‘alteration’ zones or selvedges, the subject of this study, are located adjacent to the planar pegmatites (Fig. 2) and associated parallel quartz veins. There have been brief descriptions of the alteration zones (e.g. Grew & Manton 1979; Sheraton et al. 1980; Grew 1981) but no systematic study of metasomatic processes, their absolute timing or the role they play in the geological evolution of the Napier Complex has
Fig. 2. Photograph and sketch map of sample site within the Tonagh Island orthogneiss adjacent to an early Palaeozoic pegmatite (pegmatite 29). The conspicuous ‘bleached’ alteration zone adjacent to pegmatite should be noted. The visible outer boundary of the alteration zone indicates the pyroxene-out isograd in the host orthogneiss. Series 28 sample locations as illustrated; sample 27 is located 20 m to the left of the pegmatite out of the photograph. Photographer facing east; geologist for scale; global positioning system (GPS) location 678050 37.300 S, 0508170 09.800 E (WGS 1984 reference geoid). Figure reproduced and modified from Carson et al. (2002b) with permission from Elsevier Publishing.
286
C. J. CARSON & J. J. AGUE
been conducted, in spite of their widespread occurrence. In this paper, we present petrological and geochemical information on the pegmatites and their associated metasomatic alteration zones. We highlight the important constraints that these features place on the timing of fluid influx and of upper amphibolite-facies metamorphism within the Napier Complex and the regional tectonic implications during the Early Palaeozoic. We also examine the extent of non-volatile mass transfer and comment on the nature of contrasting alkali patterns associated with this metasomatism.
Tonagh Island This study focuses on Tonagh Island, located in Amundsen Bay, Enderby Land (Figs 1 and 3). Comprising felsic orthogneisses, mafic and ultramafic granulites and subordinate lenses of Mg –Al paragneisses, the general geology of Tonagh Island was first described in detail by Harley (1985) and more recently by Osanai et al. (1999, 2001), who subdivided the island into five lithological subdivisions (Units I –V; Fig. 3). These units reflect coherent blocks of rock associations, the boundaries of which are marked by steeply north-dipping to vertical shear zones of unknown age. The great antiquity of igneous protoliths across the Napier Complex has been indicated by a number of studies (e.g. Black et al. 1986a; Harley & Black 1997). In the Tonagh Island region, Owada et al. (1994) proposed
that the mafic gneisses are the metamorphosed relics of Archaean komatiites and basalts that crystallized at 3708 + 533 Ma based on Sm–Nd whole-rock isochrons. Shiraishi et al. (1997) reported two 207Pb/206Pb zircon spot ages of c. 3280 Ma and c. 3230 Ma from a quartzofeldspathic unit within a supracrustal sequence located on the northern tip of Tonagh Island (Fig. 3); these two analyses presumably indicate the presence of Mesoarchaean zircon detritus. Carson et al. (2002a) concluded that the emplacement of an orthopyroxene-bearing felsic orthogneiss exposed on central Tonagh Island occurred at 2626 + 28 Ma. Emplacement was followed by a period of extensive zircon growth (partly characterized by elevated Th/U values) and resetting at c. 2450 –70 Ma, taken to represent the timing of UHT metamorphism on Tonagh Island. The zircon data of Shiraishi et al. (1997; Fig. 3) indicate a continuum of U –Pb ages from 2550 Ma to 2440 Ma, thought to date either a single protracted high-grade event or two events separated by c. 100 Ma. It should be stressed, however, that the timing of UHT metamorphism has been the topic of considerable debate over the past two decades. Other studies have concluded that the timing of UHT metamorphism was much earlier at c. 2850 Ma (e.g. Harley & Black 1997). More recent studies that examined REE patterns in zircon and coexisting peak garnet concluded that the age of UHT metamorphism was at least
Fig. 3. Geochronology sample localities and lithological subdivisions for Tonagh Island.
METASOMATISM OF THE NAPIER COMPLEX
c. 2510 Ma but more likely c. 2545 Ma and the previously reported ages of 2490–2450 Ma instead reflect retrograde zircon growth (e.g. Kelly & Harley 2005). Post-orogenic unmetamorphosed and undeformed felsic peraluminous pegmatites are common in the southwestern regions of the Napier Complex near the inferred Napier–Rayner boundary (Sheraton et al. 1987), and particularly in the Casey Bay region (Fig. 1; Grew 1981; Black et al. 1983a). These pegmatites decrease in abundance to the NE and east. Uranium–lead ages of c. 520 Ma (Grew 1978; Grew & Manton 1979) and an Rb–Sr isochron age of 522 + 10 Ma (Black et al. 1983a) are considered to represent the emplacement age of such pegmatites in the Casey Bay region. Carson et al. (2002b) presented U–Th–Pb data on monazite extracted from the pegmatite sample 29 from central Tonagh Island (Figs 2 and 3), and concluded that emplacement occurred at 498 + 1.7 Ma. Additional Th–Pb monazite geochronology was conducted as part of this study and will be presented below. On Tonagh Island, post-orogenic pegmatites are narrow (usually ,400 mm wide), laterally discontinuous on a scale of several hundred metres, and have been emplaced in north–south-striking subparallel sets, which dip steeply (c. 70 –808) to the east. Adjacent to these pegmatites are spectacular aureoles of upper amphibolite-facies alteration (Fig. 2). Alteration zones or selvedges are light grey in colour where they transect orthopyroxenebearing felsic orthogneiss, sharply contrasting with the light brown to buff-coloured unaltered host orthogneiss. At locations where pegmatites transect mafic or intermediate host lithologies, the alteration selvedges are dark green, reflecting development of a typical hornblende-dominated titanite þ plagioclase amphibolite-facies assemblage. The width of alteration varies from centimetre scale along narrow granitic veins to as much as c. 2 m (e.g. Fig. 2). Pegmatites and associated alteration selvedges are locally common, may be laterally continuous up to several hundred metres, and may locally affect as much as an estimated 10% of the basement lithology. Thin fractures (c. 1–10 mm wide), filled with milky vein quartz and surrounded by narrow (,10 mm) garnet þ hornblende þ epidote (+ biotite) alteration haloes, are also common on Tonagh Island. Minor fractures lacking vein quartz are also present.
Sample locations and descriptions Sample sites are located within the Unit II lithological subdivision of Osanai et al. (1999, 2001) and are shown in Figure 3. Detailed petrographic descriptions are provided below for two
287
representative examples, an orthopyroxene-bearing felsic orthogneiss and an orthopyroxene–plagioclase –quartz orthogneiss, both of which underwent alteration immediately adjacent to Early Palaeozoic pegmatites. (1) Orthopyroxene-bearing quartzo-feldspathic orthogneiss is a common lithology on Tonagh Island (Harley 1987; Osanai et al. 1999) and throughout much of the Napier Complex (e.g. Sheraton et al. 1987). Carson et al. (2002a) reported a c. 2626 Ma U –Pb zircon age from this unit on Tonagh Island, which was concluded to represent the timing of magmatic emplacement. The unit is light brown- to buff-coloured, homogeneous in appearance, with a well-developed gneissosity defined by 1– 10 mm scale discontinuous quartzofeldspathic segregations. At this sample location, the scale of the visible alteration extends c. 1000 mm from each side of the central pegmatite (Fig. 2). A series of eight samples were obtained across a traverse extending from the relatively unaltered host gneiss to immediately adjacent to pegmatite 29; in the remainder of this paper we refer to this sample set as series 28 consisting of samples 27, 30, 28/5, 28/4, 28/3, 28/2, 28/1 and 28/6 (Fig. 2) in order of increasing alteration. The orthogneiss distal from the pegmatite margin (c. 20 m; sample 27) is composed of anhedral quartz (c. 35% mode), K-feldspar (c. 14%), and plagioclase (c. 22%), and subhedral mesoperthite (Table 1). Orthopyroxene (c. 3–5%) and minor clinopyroxene (,1%) occur as dispersed subhedral ragged grains (Fig. 4a and b), rimmed by narrow, discontinuous fringes of green hornblende. Rounded magnetite –ilmenite pairs are present up to 1% mode, are ,1 mm in size, and may be partially surrounded by minor development of unoriented biotite coronas. Apatite is a common accessory phase (,1%) occurring as dispersed subhedral grains up to c. 0.5 mm in size, with minor development of narrow discontinuous epidote and light rare earth element (LREE)-enriched epidote coronas. Garnet is rare (see also Osanai et al. 1999), and in the course of this study one garnet (,1 mm diameter, with abundant quartz inclusions) was observed. Zircon is a common accessory (Table 1) and is found as small rounded to subhedral grains or euhedral elongate crystals ranging in size from 20 to 400 mm (Carson et al. 2002a, b). Immediately exterior to the visible ‘alteration zone’ (c. 1100 mm from the pegmatite margin, sample 30; Fig. 2), biotite growth is prevalent as radiating unoriented splays on magnetite–ilmenite pairs. Minor development of biotite and hornblende is present on corroded orthopyroxene. Monazite(Ce) grew on apatite, forming small anhedral grains (c. 20–30 mm) that are mantled by thin rinds of epidote and REE-enriched epidote. Quartz
288
Table 1. Mineral modal analysis of series 28 traverse Sample:
28/1
28/2
28/3
28/4
28/5
30
27
Quartz Plagioclase K-feldspar Mesoperthite*
28.4 29.8 7.0 10.3 (60.2:39.8)
29.6 33.0 7.1 10.0 (56.1:43.9)
32.0 26.0 10.5 17.9 (54.1:45.9)
32.5 27.7 9.0 17.6 (58.0:42.0)
28.7 25.3 9.1 20.3 (55.0:45.0)
28.9 20.5 12.7 26.0 (52.2:47.8)
29.5 27.6 15.0 18.3 (44.0:56.0)
32.3 19.7 16.3 22.0 (48.4:51.6)
Total K-feldspar Total plagioclase
11.1 36.0
11.5 38.6
18.7 35.7
16.4 37.9
18.2 36.5
25.1 34.1
25.2 35.7
27.7 30.3
Orthopyroxene Clinopyroxene Hornblende Biotite Magnetite Ilmenite Fe-sulphides‡
– – – 20.2 2.8† – 0.1
– – tr. 17.1 1.8 – 0.1
– – 1.0 8.4 1.7 – 0.1
– – 2.2 8.9 1.5 – 0.2
– – 3.3 9.8 2.1 0.6 tr.
– – 3.7 4.8 1.7 1.4 tr.
3.7 0.2 0.6 1.9 2.3 0.6 tr.
5.1 0.4 – 0.4 1.5 2.0 –
(Total opaque)
2.9
1.9
1.8
1.7
2.8
3.1
2.9
3.5
Apatite Epidote/REE-epidote Zircon§ Monazite Distance from pegmatite margin
1.0 0.4 0.0113 (7) – 0–10 mm
0.7 0.6 0.0168 (6) – 160 mm
0.5 0.3 0.0151 (5) – 300 mm
0.3 0.2 0.0176 (6) – 400 mm
0.6 0.2 0.0156 (5) – 550 mm
0.2 0.1 0.0120 (2) – 800 mm
0.4 tr. 0.0087 (7) tr. 1100 mm
0.3 0.1 0.0106 (6) tr. 20 m
Modal proportions were determined using the line-integration method (Brimhall 1979) in conjunction with a PC-automated digitizing petrographic microscope. tr., trace abundance. *Values in parentheses indicate volume proportion of Ca –Na feldspar: alkali-feldspar component in mesoperthite (measured by image analysis software; averaged from 12 grains per sample). †Growth of new magnetite near pegmatite. ‡Includes pyrite and pyrrhotite aggregates. §Zircon modes determined by measuring dimensions of single zircons in a number of thin sections (number in parentheses), combined with measurements of thin-section rock area dimensions.
C. J. CARSON & J. J. AGUE
28/6
METASOMATISM OF THE NAPIER COMPLEX
289
Fig. 4. Photomicrographs of altered and unaltered lithologies. (a) Least altered felsic pyroxene orthogneiss of series 28, sample 27 (opx, orthopyroxene); plane-polarized light (PPL); scale bar represents 1 mm. (b) Typical granoblastic quartzo-feldspathic matrix of sample 27 (m, mesoperthite; p, plagioclase; q, quartz); crossed polars (XPL); scale bar represents 1 mm. (c) Hornblende (hb), quartz and biotite (bi) aggregates, replacing pyroxene, sample 28/5; PPL; scale bar represents 0.6 mm. (d) Characteristic growth of anhedral epidote (ep) on apatite (ap) and as euhedral grains with biotite clusters within alteration selvedge, sample 28/1; PPL; scale bar represents 0.2 mm. (e) Extensive development of biotite in most altered sample in series 28, (28/6); no hornblende is present in this sample; PPL; scale bar represents 0.6 mm. (f) Recrystallized quartzo-feldspathic matrix in most altered sample (28/6) (compare with (b)); XPL; scale bar represents 0.8 mm. (g) Series 76 sample 76/7C, the grunerite (gu) aggregates, pseudomorphing orthopyroxene (not shown), with rinds of hornblende and biotite, should be noted (ilm, ilmenite); PPL; scale bar represents 0.7 mm. (h) Most altered sample from series 76C (sample 76/2C) showing extensive development of grunerite, hornblende, biotite and subhedral garnet (gt); PPL; scale bar represents 0.5 mm.
290
C. J. CARSON & J. J. AGUE
is characterized by extensive sub-grain development and grain boundary migration. K-feldspar and plagioclase exhibit recrystallization, reduction in grain size and serrated grain margins. The visual boundary in the field from buff-coloured to light grey orthogneiss marks the location of the disappearance of pyroxene (Fig. 2). Immediately within the visible ‘alteration zone’ (sample 28/5; Fig. 2), pyroxene is pseudomorphed by aggregates of hornblende– quartz, commonly with an outer perimeter of coarse hornblende + biotite (Fig. 4c). Biotite coronas on magnetite–ilmenite pairs are well developed, and isolated aggregates of unoriented biotite are common. With increasing alteration, biotite mode increases, whereas hornblende reaches a maximum mode of c. 4% then decreases (Table 1). Simply twinned, subhedral epidote is commonly associated with biotite aggregates (Fig. 4d). LREE-bearing epidote growth on apatite is increasingly prevalent (Fig. 4d). Monazite, either as isolated grains or rimming apatite, is not observed within the visible alteration zone. Immediately adjacent to the pegmatite, hornblende and ilmenite are absent and biotite is abundant (Fig. 4e). Plagioclase and K-feldspar are strongly recrystallized (Fig. 4f); alkali feldspar (and the alkali feldspar component of mesoperthite) steadily decreases in modal abundance with increasing alteration (Table 1). Mesoperthite lamellae show increasing resorption and ragged appearance and host submicron-scale epidote crystals along lamellae grain boundaries. Magnetite is the only oxide present; it is associated with biotite and occurs either as elongated skeletal grains or as rounded irregular grains with quartz inclusions. No deformation fabric or preferred orientation of matrix grains is present, suggesting that metasomatism and pegmatite emplacement was not accompanied by penetrative deformation. (2) Intermediate to mafic lithologies are common on Tonagh Island (Harley 1987; Osanai et al. 1999). The unit investigated comprises medium-grained (1–2 mm) plagioclase (c. 30%), orthopyroxene (c. 35–40%) and quartz (c. 35%). Accessory minerals include ilmenite–magnetite pairs (up to 5%), apatite and trace zircon (Table 2). No primary potassium-bearing phases are present. Metasomatism does not produce a clearly defined, macroscopically visible alteration zone as in the previous example (series 28), and the length scale of alteration is of the order of c. 300 mm. The sample traverse studied for this rock type is collectively referred to as series 76; sample locations are shown in Figures 3 and 5. Alteration is first apparent with replacement of orthopyroxene by grunerite (Fig. 4g); small magnetite grains (c. 20 mm) commonly define the original margins of the orthopyroxene grains. The onset of alteration was also accompanied by the development of almandine garnet mantles on grunerite. All
orthopyroxene was consumed by grunerite pseudomorphs at c. 165 mm from the pegmatite margin. Hornblende and biotite commonly rim grunerite, and both phases increase in modal abundance with increasing alteration towards the pegmatite. Apatite also increases from ,1% mode to c. 4%. In the most recrystallized region immediately adjacent to the pegmatite, garnet is found as large (1–3 mm) euhedral grains with abundant quartz inclusions, and appears in textural equilibrium with biotite, hornblende, grunerite, quartz, anhedral plagioclase and apatite (Fig. 4h). As with series 28, no strong deformation fabric or preferred orientation of matrix grains is present; metasomatism and pegmatite emplacement was not accompanied by significant penetrative deformation. (3) Mineralogy of the pegmatites associated with alteration, 29 and 76, (Figs 2 and 5), is similar. Both are coarse-grained (1–5 cm) monzogranites dominated by albitic plagioclase intergrown with lesser amounts of rounded quartz and subhedral K-feldspar. K-feldspar invariably shows exsolution of almost pure albite lamellae that are 5–10 mm wide. Biotite and muscovite are present in both pegmatites. Biotite is the dominant mica in pegmatite 29, occurring as scattered isolated grains around 1–2 mm in size, and muscovite is present as small unoriented aggregates and isolated grains. In contrast, subhedral muscovite is the dominant mica in pegmatite 76. Biotite is present as minor grains, commonly on or near the margins of muscovite. Almandine–spessartine garnet (,1 mm diameter) is present in both pegmatites, and has low modal abundance (,1%). Monazite, Fe-oxides and Fe–Cu sulphides are present as accessory phases. Zircon was not observed either in thin section or during accessory heavy mineral separation procedures. Although not observed in these samples, similar pegmatites in the sample vicinity contain large (c. 2–3 cm) euhedral greenish beryl and small clusters of dumortierite fibres (E. S. Grew, pers. comm.).
Analytical techniques Bulk-rock geochemical traverses were conducted across the alteration zones described above. Major elements were analysed on fused discs, trace elements (Ba, Rb, Sr, Nb and Zr) on pressed powder discs, and REE (including Y, Th, U and Pb) by solution inductively coupled plasma mass spectrometry (ICP-MS). Analyses were conducted by X-Ray Assay Laboratories (XRAL), Don Mills, Ontario, Canada. Rock grain densities were measured on pulverized samples using a Micromeritics manual gas pycnometer at Yale University. Modal determinations were obtained using the lineintegration method (e.g. Brimhall 1979; Ague
Table 2. Modal abundance of minerals in plagioclase– orthopyroxene host orthogneiss, series 76C traverse 76/2C
76/3C
76/4C
76/5C
76/6C
76/7C
76/8C
76/9C
76/10C
Protolith*
Quartz Plagioclase Garnet Orthopyroxene Grunerite Hornblende Biotite Magnetite Ilmenite Apatite Epidote/REE-epidote Zircon Distance from pegmatite (mm)
12.2 15.6 9.6 – 15.7 17.9 22.0 0.7 2.2 4.3 0.3 – 0–7
15.7 19.2 12.2 – 18.7 12.8 11.8 0.4 4.5 4.9 0.2 tr. 28
15.9 23.8 11.8 – 22.8 10.0 9.4 0.4 2.8 3.0 0.1 – 50
12.0 27.1 16.7 – 25.4 7.3 3.1 0.5 6.1 1.3 0.1 – 73
12.2 29.8 7.2 – 30.5 9.0 3.3 2.0 4.7 1.2 – – 98
22.2 27.3 2.7 – 34.6 5.7 1.2 1.2 4.1 0.4 – – 122
19.8 41.3 2.4 – 23.4 5.1 1.9 3.9 1.5 0.5 – tr. 145
17.1 40.9 0.9 – 28.9 3.8 2.3 1.5 4.0 1.0 – – 165
13.5 46.1 4.6 1.2 24.1 3.5 2.0 1.2 2.5 0.4 – tr. 187
13.5 48.4 0.0 37.2 0.0 0.0 0.0 1 2 0.4 0.0 ? ?
Modal proportions were determined using the line-integration method (Brimhall 1979) in conjunction with a PC-automated digitizing petrographic microscope. tr., trace abundance. *Protolith values were approximated by replacing the volume percent of Fe –Mg-bearing alteration products (garnet, grunerite, hornblende and biotite) with that of orthopyroxene.
METASOMATISM OF THE NAPIER COMPLEX
Sample:
291
292
C. J. CARSON & J. J. AGUE
Fig. 5. Sample numbers and locations for series 76; from hand-sample specimens. Two sample traverses were done. (a) Series 76B for XRF analysis (Table 10); (b) series 76C for thin sections (locations indicated by the outlined boxes) for modal counting (Table 2) and electron probe microanalysis (Table 3 –7). Indicated measurements (in mm) are taken from the main pegmatite margin in both cases.
1994b) in conjunction with a PC-automated digitizing petrographic microscope. Pressure–temperature estimates were calculated using THERMOCALC v3.0 (Powell & Holland 1994). Mineral compositions were determined using the JEOL JXA-8600 electron microprobe at Yale University, employing wavelength-dispersive spectrometers, natural and synthetic standards, and frz matrix corrections. Analytical conditions were 15 kV accelerating voltage, 15 nA beam current, and a 10 mm beam spot for muscovite and biotite and a 5 mm beam spot for garnet and feldspars. Mineral compositions are presented in Tables 3–7. Primary compositions of pegmatite exsolved feldspars were determined by computer image analysis of back-scattered electron exsolved feldspar images. The relative surface area proportion of exsolution lamellae and host were thus determined from the average of as many as 10 grains; the compositions of feldspar components were determined by electron probe analysis. Using known molar volumes (see Appendix), the molar proportion of each component was then
calculated (Table 3). The ‘reintegrated’ pre-solution feldspar composition was then calculated using the molar proportions and exsolution lamellae and host composition information. Monazite-(Ce) from pegmatite 76 was separated using standard heavy liquid and magnetic techniques. Grains were large (c. 200 mm), clear, glassy and unfractured. Selected grains were mounted in epoxy, with reference monazite standard 554, sectioned approximately in half, polished and Au-coated. Backscattered electron imaging was conducted to facilitate analysis site selection. Ion microprobe analyses were conducted using the Cameca IMS 1270, located at the Department of Earth and Space Sciences, University of California, Los Angeles, USA. Monazite standard 554 (Harrison et al. 1995, 1999) was analysed together with the unknowns. 232Th– 208Pb analytical protocols have also described by Harrison et al. (1995, 1999). Isotopic ratios were corrected, where necessary, for contemporary common lead values using measured 204 Pb and the crustal common Pb values of Stacey & Kramers (1975). Radiogenic isotope data were
Table 3. Representative feldspar compositions Oxides
K-feldspar (wall-rock) 27*
27†
64.23 64.90 18.27 18.41 0.00 0.00 0.03 0.04 0.75 0.59 0.99 1.19 15.10 14.86 99.37 100.00
Si Al Fe Ca Ba Na K Sum XOr XAn XAb aOr k aAn aAb
2.99 1.00 0.00 0.00 0.01 0.09 0.90 5.00 89.8 0.2 9.0 0.91 – 0.37
27†
Pegmatite 76
28/5* 28/6* 28/6† 76/10 76/2 K-feldspar Ab exsolution reintegrated‡ Plagioclase K-feldspar Ab exsolution reintegrated§ Plagioclase
64.06 64.51 60.15 60.61 60.35 62.56 62.61 59.02 60.93 18.23 18.57 24.39 24.59 24.29 23.02 23.01 25.54 24.32 0.00 0.00 0.01 0.02 0.01 0.07 0.01 0.13 0.23 0.04 0.01 6.25 6.49 6.04 4.62 4.66 7.73 5.95 0.85 0.81 0.09 0.00 0.00 0.00 0.02 0.01 0.03 0.78 0.71 8.07 8.04 8.06 8.97 9.04 7.21 8.25 15.53 15.46 0.19 0.09 0.14 0.08 0.12 0.17 0.48 99.48 100.07 99.14 99.84 98.88 99.31 99.46 99.80 99.76
3.00 2.99 1.00 1.00 0.00 0.00 0.00 0.00 0.01 0.02 0.11 0.07 0.88 0.93 4.99 5.01 87.5 92.5 0.2 0.2 10.6 7.0 0.89 0.93 – – 0.42 0.30
Pegmatite 29
2.99 2.70 2.70 2.71 2.79 2.79 2.64 2.72 1.01 1.29 1.29 1.29 1.21 1.21 1.35 1.28 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.01 0.00 0.30 0.31 0.29 0.22 0.22 0.37 0.28 0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.06 0.70 0.70 0.70 0.77 0.78 0.63 0.71 0.91 0.01 0.01 0.01 0.01 0.01 0.01 0.00 4.99 5.01 5.00 5.00 5.00 5.00 5.00 5.00 91.3 1.1 0.5 0.8 0.5 0.7 1.0 0.3 0.1 30.1 31.0 29.1 22.1 22.2 37.1 28.4 6.4 70.3 69.5 70.2 77.4 78.0 62.6 71.3 0.92 – – – – – – – – 0.47 0.48 0.44 0.31 0.32 0.57 0.43 0.27 0.63 0.63 0.64 0.72 0.72 0.57 0.65
65.19 18.40 0.02 0.07 0.01 1.53 14.54 99.75
67.09 20.46 0.00 1.12 0.02 11.21 0.08 99.98
65.51 18.75 0.02 0.21 0.01 3.16 12.09 99.74
65.89 21.26 0.02 2.28 0.03 10.25 0.20 99.92
65.05 18.31 0.01 0.03 0.00 1.55 14.67 99.62
67.72 19.15 0.00 0.05 0.01 11.91 0.10 98.95
65.33 18.40 0.01 0.04 0.00 2.64 13.14 99.56
66.02 21.17 0.00 2.11 0.00 10.33 0.39 100.01
3.00 1.00 0.00 0.00 0.00 0.14 0.85 4.99 85.4 0.4 13.6 0.87 – 0.52
2.94 1.06 0.00 0.05 0.00 0.95 0.01 5.01 0.5 5.2 95.3 – – –
2.99 1.01 0.00 0.01 0.00 0.28 0.70 5.00 70.4 1.0 28.0 0.79 – 0.81
2.90 1.10 0.00 0.11 0.00 0.87 0.01 4.99 1.1 10.7 87.4 – 0.12 0.86
3.00 1.00 0.00 0.00 0.00 0.14 0.86 5.00 86.4 0.2 13.9 0.88 – 0.55
2.99 1.00 0.00 0.00 0.00 1.02 0.01 5.02 0.6 0.3 100.0 – – –
3.00 1.00 0.00 0.00 0.00 0.24 0.77 5.00 77.0 0.2 23.5 0.81 – 0.79
2.09 1.10 0.00 0.10 0.00 0.88 0.02 5.00 2.2 10.0 88.0 – 0.12 0.86
METASOMATISM OF THE NAPIER COMPLEX
SiO2 Al2O3 FeO CaO BaO Na2O K2O Total
28/6* 28/6† 27*
Plagioclase (wall-rock)
Oxygens ¼ 8. *Isolated matrix grain. †Mesoperthite lamellae component. ‡83.1 mol% K-feldspar component;16.6 mol% albite exsolution; reintegrated indicates reintegration of exsolved and host feldspar components. §89.5 mol% K-feldspar component;10.5 mol% albite exsolution; reintegrated indicates reintegration of exsolved and host feldspar components. k Feldspar activities after Fuhrman & Lindsley (1988); it should be noted that end-members with low mole fractions will have large errors and have been disregarded.
293
294
C. J. CARSON & J. J. AGUE
Table 4. Representative pyroxene compositions Oxides
opx 27
opx 27
cpx 27
opx 76/10
opx 76/10
SiO2 TiO2 Al2O3 FeO MnO MgO CaO Na2O Total
51.36 0.07 0.98 29.49 0.63 16.86 0.52 0.00 99.93
51.51 0.02 0.88 29.99 0.76 16.48 0.42 0.00 100.07
52.35 0.14 1.87 11.63 0.23 11.37 21.53 0.69 99.81
50.74 0.05 1.27 33.38 0.52 13.34 0.72 0.04 100.06
50.28 0.05 1.30 33.50 0.49 13.41 0.83 0.04 99.91
1.98 0.00 0.05 0.95 0.02 0.97 0.02 0.00 3.99
1.99 0.00 0.04 0.97 0.03 0.95 0.02 0.00 3.99
1.98 0.00 0.08 0.37 0.01 0.64 0.87 0.05 4.00
1.99 0.00 0.06 1.10 0.02 0.78 0.03 0.00 3.98
1.98 0.00 0.06 1.10 0.02 0.79 0.04 0.00 3.99
Si Ti Al Fe Mn Mg Ca Na Sum XWo* XEn† XFs‡
1.1 49.9 49.0
0.9 49.0 50.1
46.4 34.1 19.6
1.6 40.9 57.5
1.8 40.9 57.3
Oxygens ¼ 6. *XWo, mole proportion of wollastonite 100. †XEn, mole proportion of enstatite 100. ‡XFs, mole proportion of ferrosilite 100.
processed using ISOPLOT version 2.3 (Ludwig 1999). Monazite isotope data from pegmatite 76 are presented in Table 8. The errors listed in Table 8 are 1s, the weighted mean ages quoted in the text are 2s.
Results Pressure – temperature estimates For H2O-saturated conditions, the average pressure– temperature estimates for the pegmatite 76 assemblage (garnet, biotite, muscovite, K-feldspar, plagioclase and quartz; Tables 3–7) is 8.1 kbar and 681 8C. Fluid-absent calculations return 8.2 kbar and 678 8C. These results are identical at the 2s level (minimum uncertainties +c. 34 8C, +c. 1.6 kbar), suggesting that the infiltrating fluid was dominated by H2O. These results are in excellent agreement with P–T estimates obtained from pegmatite 29 (Carson et al. 2002b) that yield H2O-saturated conditions of 8.1 kbar and 684 8C and fluid-absent results of 8.0 kbar and 666 8C. Carson et al. (2002b) also estimated the HCl and HF contents of the infiltrating fluid, based on biotite F and Cl contents (Zhu & Sverjensky 1992), and concluded that the fluid was characterized by low halogen abundances. These results indicate that the crystallization P–T
Table 5. Representative garnet compositions Oxides
SiO2 TiO2 Al2O3 FeO MnO MgO CaO Total Si Ti Al Fe Mn Mg Ca Sum XAlm XPy XGr XSpss Oxygens ¼ 12. *Next to hornblende. †Next to biotite.
Sample 76/2C
Sample 76/10
76 pegmatite
rim*
rim†
core
rim
core
rim
37.41 0.08 21.30 32.49 0.94 3.71 3.93 99.85
37.26 0.00 21.57 32.79 0.80 3.38 4.18 99.97
37.17 0.09 21.15 32.12 1.11 3.23 4.84 99.72
37.35 0.06 21.21 31.66 0.78 3.49 4.93 99.48
37.04 0.02 21.05 33.03 0.81 3.05 4.44 99.43
36.71 0.00 20.72 31.22 7.65 1.16 2.16 99.62
2.99 0.01 2.00 2.17 0.06 0.44 0.34 8.01
2.98 0.00 2.03 2.19 0.05 0.40 0.36 8.01
2.98 0.01 2.00 2.15 0.08 0.39 0.42 8.02
2.99 0.00 2.00 2.12 0.05 0.42 0.42 8.01
2.99 0.00 2.00 2.23 0.06 0.37 0.38 8.02
3.00 0.00 2.00 2.14 0.53 0.14 0.19 8.00
72.1 14.7 11.2 2.1
72.9 13.4 11.9 1.8
71.0 12.7 13.7 2.5
70.4 13.8 14.0 1.8
73.4 12.1 12.7 1.8
71.3 4.7 6.3 17.7
METASOMATISM OF THE NAPIER COMPLEX
295
Table 6. Representative apatite and clinozoisite –allanite compositions Oxides
Apatite*
Clinozoisite – allanite
28/6
27
76/2
76/10
28/3‡
28/6§
28/6§
76/2
76/2§
0.00 0.02 nd nd nd 55.75 42.08 0.16 0.04 0.06 0.04 0.01 0.01 0.00 0.04 0.00 2.52 0.05 21.07 99.70
0.16 0.05 nd nd nd 54.56 40.52 0.30 0.08 0.18 0.01 0.19 0.03 0.06 0.01 0.00 2.97 0.00 21.26 97.92
0.00 0.06 nd nd nd 55.29 41.55 0.11 0.04 0.10 0.06 0.02 0.04 0.05 0.06 0.00 2.70 0.04 21.15 98.93
0.06 0.01 nd nd nd 54.42 40.07 0.34 0.00 0.08 0.07 0.07 0.03 0.09 0.06 0.00 2.15 0.05 20.92 96.57
37.97 23.33 13.24 0.25 0.03 22.99 nd 0.00 0.00 0.00 0.00 0.20 0.00 0.00 nd nd nd nd – 98.01
32.80 17.41 14.18 0.42 0.47 12.03 nd 0.00 6.42 11.62 1.20 2.43 0.00 0.00 nd nd nd nd – 98.98
35.07 20.48 12.72 0.33 0.29 16.62 nd 0.33 3.62 6.39 0.62 1.85 0.00 0.00 nd nd nd nd – 98.31
33.83 19.46 13.47 0.05 0.62 13.10 nd 0.54 2.94 7.38 1.09 4.51 0.57 0.40 nd nd nd nd – 97.96
33.56 19.44 12.88 0.02 0.69 13.62 nd 0.52 3.07 7.83 0.84 4.49 0.64 0.33 nd nd nd nd – 97.92
Si Al Fe3þ Fe2þ k Mn Mg Ca P Y La Ce Pr Nd Sm Gd U Th Sum
0.00 0.00 – – – – 4.96 3.01 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 7.98
0.01 0.01 – – – – 5.03 2.95 0.01 0.00 0.01 0.00 0.01 0.00 0.00 0.00 0.00 8.03
0.00 0.00 – – – – 5.02 2.98 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 8.01
0.01 0.00 – – – – 5.07 2.95 0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 8.05
3.02 2.19 0.79 – 0.17 0.00 1.96 – 0.00 0.00 0.00 0.00 0.01 0.00 0.00 – – 7.99
2.97 1.86 0.25 0.79 0.03 0.06 1.17 – 0.00 0.21 0.39 0.04 0.08 0.00 0.00 – – 7.77
3.03 2.06 0.41 0.41 0.02 0.04 1.52 – 0.02 0.11 0.20 0.02 0.06 0.00 0.00 – – 7.86
2.98 2.02 0.45 0.44 0.00 0.08 1.24 – 0.03 0.10 0.24 0.04 0.14 0.02 0.01 – – 7.78
2.97 2.03 0.29 0.57 0.00 0.09 1.29 – 0.02 0.10 0.25 0.03 0.14 0.02 0.01 – – 7.81
F Cl
0.72 0.01
0.81 0.01
0.73 0.01
0.59 0.01
– –
– –
– –
– –
– –
SiO2 Al2O3 Fe2O3† MnO MgO CaO P2O5 Y2O3 La2O3 Ce2O3 Pr2O3 Nd2O3 Sm2O3 Gd2O3 ThO2 UO2 F Cl (O5 5F, Cl) Total
Oxygens ¼ 12.5; nd, not determined; – , not calculated. *Apatite analyses are average of five analyses from single grain. †All Fe Calculated as Fe2O3. ‡Large (c. 0.5 mm) twinned REE-poor grain associated with biotite clusters. §Allanite mantle on apatite. k Calculated from Fe3þ on the basis of the coupled allanite substitution Ca2þ þ Fe3þ ¼ (Y)REE3þ þ Fe2þ.
conditions of felsic pegmatites were c. 8 kbar at c. 670–680 8C. Although satisfactory multi-reaction thermobarometry results could not be obtained on wall-rock assemblages, hornblende–plagioclase temperatures (Holland & Blundy 1994) are 669–683 8C for various 28 series samples and 679–681 8C for 76/ 2C (at 8 kbar). We conclude that the local wall-rock temperature, at least close to the pegmatite, was
c. 670–680 8C (consistent with the P–T results obtained from the pegmatites), although these estimates may not necessarily reflect the regional ambient temperature conditions. Although wallrock temperatures of c. 670–680 8C under fluidsaturated conditions would be above the wet solidus for felsic systems, and minor partial melting of series 28 wall-rock might have been expected (e.g. Thompson & Tracy 1979), our
296
Table 7. Representative mica and amphibole compositions Oxides
Hornblende
Grunerite
Muscovite
76/10C
76/2C
76 pegmatite
28/6
28/5
28/5*
28/5†
28/3
76/10C
76/2C
76/10C
76/2C
76 pegmatite
36.46 3.58 15.03 17.90 0.11 12.31 0.06 0.60 0.09 9.10 0.47 0.02 3.94 20.20 99.46
37.16 2.87 15.07 18.96 0.09 11.85 0.07 0.24 0.16 9.02 0.43 0.02 3.95 20.19 99.69
33.81 0.58 19.94 27.94 0.24 3.94 0.00 0.00 0.04 8.96 0.08 0.07 3.81 20.05 99.36
35.13 4.33 14.87 24.12 0.48 7.22 0.06 2.59 0.04 9.24 0.19 0.04 3.83 20.09 99.71
35.19 4.25 14.33 24.76 0.45 7.61 0.08 0.33 0.11 9.01 0.19 0.02 3.82 20.09 100.06
46.23 0.38 8.2 19.29 0.70 9.85 11.20 0.14 0.77 0.65 0.22 0.01 2.08 20.09 99.65
41.35 1.07 12.21 21.63 0.67 6.88 11.15 0.14 1.38 1.37 0.08 0.01 2.03 20.04 99.96
41.38 0.75 12.3 22.25 0.53 6.64 11.30 0.00 1.18 1.42 0.08 0.02 1.94 20.04 99.76
42.45 1.00 13.64 19.45 0.21 8.14 10.75 0.08 1.35 0.70 0.15 0.03 2.07 20.07 99.95
43.63 0.88 12.25 18.62 0.13 9.11 11.10 0.17 1.37 0.41 0.24 0.00 2.08 20.10 99.89
53.09 0.24 0.86 27.12 0.32 15.24 0.53 0.16 0.07 0.05 0.22 0.03 2.10 20.10 99.92
54.39 0.24 1.00 25.55 0.24 16.17 0.85 0.04 0.11 0.02 0.19 0.01 2.13 20.08 100.86
46.70 0.07 33.60 3.78 0.01 0.46 0.01 0.00 0.55 10.13 0.04 0.00 4.47 20.02 99.80
Oxygens
22
22
22
22
22
23
23
23
23
23
23
23
22
Si Ti Al Fe Mn Mg Ca Ba Na K Sum
5.11 0.38 2.48 2.10 0.01 2.57 0.01 0.03 0.02 1.63 14.36
5.19 0.30 2.48 2.22 0.01 2.47 0.01 0.01 0.04 1.61 14.35
4.91 0.06 3.41 3.40 0.03 0.85 0.00 0.00 0.01 1.66 14.33
5.05 0.47 2.52 2.90 0.06 1.55 0.01 0.02 0.01 1.70 14.30
5.06 0.46 2.43 2.98 0.06 1.63 0.01 0.02 0.03 1.65 14.33
6.70 0.04 1.41 2.34 0.09 2.13 1.74 0.01 0.22 0.12 14.77
6.11 0.12 2.13 2.67 0.09 1.52 1.77 0.01 0.40 0.26 15.06
6.11 0.08 2.15 2.76 0.07 1.47 1.80 0.00 0.34 0.27 15.07
6.14 0.11 2.33 2.35 0.03 1.76 1.67 0.01 0.38 0.13 14.88
6.29 0.10 2.08 2.24 0.02 1.96 1.71 0.01 0.38 0.08 14.86
7.58 0.03 0.14 3.24 0.04 3.24 0.08 0.01 0.02 0.01 14.39
7.61 0.03 0.17 2.99 0.03 3.37 0.13 0.00 0.03 0.00 14.36
5.75 0.01 4.88 0.39 0.00 0.08 0.00 0.00 0.13 1.59 12.84
0.21 0.01 3.69
0.19 0.00 3.68
0.04 0.02 3.69
0.09 0.01 3.65
0.09 0.00 3.66
0.10 0.00 2.01
0.04 0.00 2.00
0.04 0.00 1.92
0.07 0.01 2.00
0.11 0.00 2.00
0.10 0.01 2.00
0.08 0.00 1.98
0.02 0.00 3.67
F Cl H‡ XFe (100)
44.9
47.3
79.9
An Fe given as FeO. *Hornblende þ quartz rims around pseudomorphed orthopyroxene. †Matrix hornblende. ‡Calculated assuming F þ Cl þ OH site occupancy is filled.
65.2
64.6
52.3
63.8
65.3
57.3
53.4
50.0
47.0
–
C. J. CARSON & J. J. AGUE
SiO2 TiO2 Al2O3 FeO MnO MgO CaO BaO Na2O K2O F Cl H2O‡ (O5 5F, Cl) Total
Biotite
METASOMATISM OF THE NAPIER COMPLEX
297
Table 8. Pegmatite 76 monazite Th–Pb isotope results Grain_spot
208
Pb*/232Th
+
% radiogenic 208
gr10_sp2 gr8_sp2 gr5_sp2 gr3_sp2 gr8_sp1 gr7_sp1 gr9_sp1 gr10_sp1 gr3_sp1 gr7_sp2 gr5_sp1 gr2_sp1 gr9_sp2 gr1_sp3 gr1_sp1 gr2_sp2 gr1_sp2 gr7_sp3
0.02486 0.02524 0.02598 0.02603 0.02626 0.02633 0.02642 0.02649 0.02653 0.02664 0.02673 0.02684 0.02693 0.02719 0.02723 0.02763 0.02789 0.02854
0.00039 0.00025 0.00024 0.00022 0.00027 0.00023 0.00026 0.00029 0.00024 0.00024 0.00028 0.00027 0.00033 0.00035 0.00028 0.00027 0.00026 0.00026
Pb
99.7 99.6 99.5 99.5 99.6 99.6 99.6 99.7 99.6 99.6 99.6 99.6 99.7 99.7 99.8 99.7 99.6 99.6
Age (Ma) 208
Pb/
232
Th
496.3 503.8 518.5 519.4 523.9 525.3 527.1 528.5 529.3 531.4 533.2 535.4 537.1 542.2 542.9 550.8 555.9 568.8
+† 7.6 5.0 4.6 4.4 5.3 4.6 5.2 5.8 4.8 4.7 5.6 5.3 6.5 6.9 5.6 5.4 5.1 5.1
*Correction for common Pb made using the measured 204Pb/208Pb ratio. †Uncertainties listed at the 1s level.
geochemical results (discussed below) indicate that this melt fraction was not extracted. Minor partial melting of the intermediate wall-rock (series 76) would be expected under these conditions.
Bulk-rock geochemistry The traverse geochemical data are presented, together with the sampling distance from pegmatite margin, in Tables 9 and 10 and are illustrated in Figures 6 and 7 (28 and 76 sample series, respectively). Although detailed discussion of the techniques of assessing mass change in geochemical systems is beyond the scope of this paper, the reader is directed to Ague & van Haren (1996; and references therein) and Ague (2003) for a complete discussion. We have used concentration ratio discrimination diagrams as a primary means to identify the relative enrichment or depletion of elements as a result of metasomatism between the precursor host orthogneiss and within the alteration zones (Ague 2003). This ratio represents the concentration of species of interest within the altered 0 ) relative to the concentration of selvedge (Cspecies that species within the reference precursor (C0species) according to the equation concentration ratio ¼
0 Cspecies 0 Cspecies
:
(1)
If no mass change has occurred, clearly, the ratio for all species will be unity. For an ideal example, a
number of species might undergo minimal mass transfer, permitting the calculation of a mean reference concentration ratio from such relatively immobile species, and against which the degree of mass transfer of other, more mobile, elements can be assessed (Ague & van Haren 1996). If the altered rock has lost mass overall, then the concentration ratio for these reference species will be .1; if mass has been gained it will be ,1 (see Ague 2003). Mobile species that have been added to the rock have concentration ratios greater than the reference ratio, and those that are lost from the rock have ratios less than the reference ratio (Ague 2003). In Figures 6 and 7 we have chosen the mean of the concentration ratios of Ti and Al in the altered selvedge to a relatively unaltered precursor as our geochemical reference frame. Results for a Zr reference frame are very similar. We also present quantitative mass changes of major species, including an estimate of total mass change during alteration, using either Ti or Al as ‘immobile’ reference frames (e.g. Ague 1994a) and these results are tabulated in Table 11. The determination of these values was calculated using ‘relatively’ immobile reference species according to the relationship
fractional mass change of mobile species ( j) ! o Cj0 Ci 1 ¼ Ci0 Cjo
(2)
298
Table 9. Whole-rock composition of pegmatite (29) and adjacent felsic orthogneiss (series 28) Pegmatite
Orthopyroxene-bearing felsic orthogneiss (host gneiss) 28/6
28/1
28/2
28/3
28/4
28/5
SiO2 TiO2 Al2O3 Fe2O3 MnO MgO CaO Na2O K2O P2O5 LOI
66.55 0.03 19.35 0.81 0.07 0.02 1.61 7.03 4.11 0.03 0.28
71.95 0.34 14.15 3.08 0.05 0.79 2.32 4.09 2.78 0.13 0.45
71.00 0.40 14.35 3.53 0.05 0.97 2.76 3.84 2.69 0.14 0.38
71.53 0.36 13.90 3.07 0.04 0.86 2.68 3.36 3.33 0.12 0.27
71.10 0.38 14.35 3.22 0.04 0.87 2.78 3.45 3.54 0.12 0.38
71.40 0.34 14.25 3.01 0.04 0.81 2.67 3.54 3.62 0.12 0.28
71.65 0.32 14.20 2.76 0.04 0.90 2.73 3.39 3.69 0.10 0.20
73.45 0.23 13.65 1.85 0.03 0.63 2.65 3.25 3.53 0.07 0.55
72.00 0.29 13.70 2.42 0.03 0.78 2.72 3.17 3.68 0.09 0.85
71.08 0.44 14.05 3.19 0.04 0.90 2.86 3.29 3.67 0.14 0.32
Total
99.86
100.11
100.08
99.53
100.22
100.06
99.97
99.87
99.90
99.97
Trace elements* Ba Nb Rb Sr Y Zr Pb (2) Y† (1) La (0.1) Ce (0.1) Pr (0.2) Nd (0.1)
66 16 169 41 21 32 130 28 16.3 30.0 3.6 14.6
887 17 133 192 14 117 30 23.0 61.8 102.0 10.2 33.5
1020 8 113 221 18 156 18 17.0 54.4 88.3 8.5 29.1
978 5 79 219 14 134 15 17.0 54.8 88.4 8.6 29.3
1050 5 76 231 14 145 12 17.0 52.7 86.9 8.3 28.7
1075 5 69 230 15 138 14 17.0 56.0 90.3 8.8 30.3
1119 4 75 238 11 130 16 13.0 46.8 74.0 7.0 24.2
30ave
1060 3 72 214 4 102 – – – – – –
30B
1010 3 74 208 5 115 13 7.0 46.2 67.5 5.9 18.7
27 (unaltered)
1355 6 69 257 13 167 14 15.0 56.2 90.0 8.7 30.0
C. J. CARSON & J. J. AGUE
29
6.0 0.26 5.8 1.0 5.7 0.98 2.8 0.4 2.9 0.43 11.6 16.6 – – 1.02 100.9 – 0
6.0 1.3 5.8 0.7 4.1 0.72 2.1 0.3 1.8 0.26 2.4 1.5 bd 14.9
4.9 1.4 4.8 0.6 3.5 0.64 1.8 0.2 1.5 0.20 1.7 0.4 bd 1.5
4.9 1.4 4.7 0.6 3.3 0.6 1.6 0.2 1.3 0.19 1.6 0.6 bd 1.0
4.9 1.5 4.7 0.5 3.2 0.57 1.6 0.2 1.4 0.17 1.5 0.9 bd 0.9
5.2 1.5 4.8 0.6 3.3 0.59 1.6 0.2 1.3 0.17 1.5 1.0 bd 0.8
3.7 1.4 3.9 0.4 2.5 0.43 1.2 0.2 1.1 0.13 1.1 0.7 bd 0.9
– – – – – – – – – – – – – –
86.8 2.699 10
99.8 2.671 160
175.0 2.680 300
193.3 2.692 400
217.8 2.667 550
204.2 2.675 800
203.5 – 1100
2.5 1.2 2.3 0.2 1.4 0.25 0.7 0.1 0.7 0.08 2.3 0.4 bd 1.1 206.4 2.717 1100
4.7 1.7 4.2 0.5 2.9 0.55 1.6 0.2 1.4 0.21 5.8 0.4 bd 1.0 220.7 2.690 20000
*Via XRF pressed pellet. †REE, Y, Th and U analyses via ICP-MS; B and Be via ICP-MS. Detection limits for trace elements are given in parenthesis as ppm. All major elements are an average of two whole-rock samples, except for 28/2, which is average of three. All trace elements determined on a single representative sample. All Fe as Fe2O3. LOI, loss on ignition; –, not analysed.
METASOMATISM OF THE NAPIER COMPLEX
Sm (0.1) Eu (0.05) Gd (0.1) Tb (0.1) Dy (0.1) Ho (0.05) Er (0.1) Tm (0.1) Yb (0.1) Lu (0.05) Th (0.1) U (0.1) B (10) Be (0.5) Alumina Index K/Rb Density (g cm23) Distance (mm)
299
300
Table 10. Whole-rock composition of pegmatite (76) and orthopyroxene orthogneiss host (series 76) Pegmatite
Plagioclase – orthopyroxene orthogneiss (B series) 76/10B
76/1B
76/2B
76/3B
76/4B
76/5B
76/6B
76/7B
76/9B
76/8B
SiO2 TiO2 Al2O3 Fe2O3 MnO MgO CaO Na2O K2O P2O5 LOI
75.8 0.02 14.6 0.64 0.06 0.02 1.41 6.49 0.73 0.03 0.35
50.1 2.09 12.7 17.50 0.18 5.76 4.30 0.98 3.13 1.30 0.35
50.4 2.18 12.9 18.40 0.19 6.00 4.90 1.91 1.51 0.84 0.10
51.2 2.14 13.1 18.30 0.19 5.95 4.83 2.41 0.76 0.71 20.30
53.8 1.99 13.1 16.80 0.17 5.62 4.58 2.99 0.39 0.42 20.15
55.3 1.83 13.0 15.60 0.16 5.31 4.59 3.41 0.25 0.38 20.30
57.3 1.62 13.3 13.50 0.14 4.56 4.73 3.49 0.28 0.37 20.10
57.4 1.82 13.2 13.70 0.14 4.62 4.77 3.63 0.22 0.36 ,0.01
56.9 1.87 13.5 13.70 0.14 4.64 4.87 3.61 0.22 0.39 20.10
55.1 1.96 13.3 15.00 0.15 5.20 5.01 3.41 0.24 0.38 20.10
54.3 2.02 13.8 15.30 0.16 5.51 5.29 3.39 0.25 0.41 20.40
Total
100.20
98.50
99.20
99.30
99.80
99.70
99.20
99.90
99.80
99.70
100.10
Trace elements (ppm)* Ba Nb Rb Sr Y Zr Pb
28 12 50 23 34 45 86
Y† (1) La (0.1) Ce (0.1) Pr (0.2)
32 9.8 17.1 1.9
218 27 214 49 187 105 24 – – – –
161 21 64 92 151 126 17
96 21 30 114 124 116 14
191 73.1 183.0 24.0
110 49.3 117.0 15.3
71 17 14 114 83 113 13 – – – –
88 18 7 118 70 112 16 – – – –
156 10 7 128 56 72 15 – – – –
128 12 4 130 57 105 11 – – – –
158 13 3 135 58 97 15 55 23.2 51.7 6.5
147 13 6 141 58 117 14 59 24.4 53.8 6.8
118 21 5 153 56 162 12 58 25.3 54.1 6.6
C. J. CARSON & J. J. AGUE
76
Alumina Index K/Rb Density (g cm23) Distance (mm)
8.4 3.8 0.25 4.7 0.9 5.8 1.1 3.1 0.4 2.7 0.40 8.3 6.6
– – – – – – – – – – – – –
1.18 60.6 – –
60.7 – 4
118.0 34.8 3.31 38.5 6.3 34.9 6.67 16.8 2.1 11.9 1.71 8.4 2.2 97.9 3.119 20
63.6 20.3 2.95 23.9 3.8 21.1 4.07 11.1 1.5 8.3 1.20 5.0 1.3
– – – – – – – – – – – – –
– – – – – – – – – – – – –
– – – – – – – – – – – – –
– – – – – – – – – – – – –
29.5 8.5 1.95 9.4 1.5 10.0 2.02 5.5 0.7 5 0.80 2.0 0.6
29.9 9.1 1.97 10.1 1.7 11.0 2.14 6.2 0.9 5.2 0.81 2.4 1.5
96.9 – 40
115.6 – 60
148.2 3.097 80
166.0 3.066 100
228.3 2.988 125
304.4 2.983 150
262.9 2.950 177
31.3 8.6 2.09 10.3 1.6 10.3 2.00 5.9 0.8 5.2 0.81 2.3 1.0 340.4 – 205
*Via XRF pressed pellet. †REE, Y, Th and U analyses via ICP-MS. Detection limits are given in parentheses as ppm. All Fe as Fe2O3. Sample locations for series 76B with respect to pegmatite 76 are shown in Figure 5. Negative LOI for the majority of the analyses here is interpreted to indicate a relatively high FeO/Fe2O3 and low volatile abundance. Oxidation of the FeO during analysis caused net mass gain, resulting in a negative LOI. – , not analysed.
METASOMATISM OF THE NAPIER COMPLEX
Nd (0.1) Sm (0.1) Eu (0.05) Gd (0.1) Tb (0.1) Dy (0.1) Ho (0.05) Er (0.1) Tm (0.1) Yb (0.1) Lu (0.05) Th (0.1) U (0.1)
301
302
C. J. CARSON & J. J. AGUE
Fig. 6. (a) Concentration ratio diagram for series 28. The horizontal dashed line represents the mean concentration ratio of Ti and Al; *, concentration ratio based on whole-rock geochemistry; W, concentration ratio calculated for K, Na and Ca based on modes, mineral compositions and molar volumes of major phases. The horizontal axis indicates the elements analysed, the vertical axis is the concentration ratio defined as the selected altered sample relative to a reference precursor, which is indicated in the relevant diagrams. Filled circles connected by a continuous line are the REE (note the Eu ‘anomaly’ and the slight preferential enrichment of the HREE). (b) La/Lu v. distance from pegmatite, illustrating the progressive enrichment of the HREE over LREE with increasing alteration of wall-rock in series 28. The limit of the visible alteration selvedge (Fig. 2) is indicated by the vertical dashed line. (c) Sm/Eu v. distance from pegmatite, illustrating the progressive development of the Eu ‘anomaly’ with increasing alteration (normalized so that Sm/Eu for sample 28/5 ¼ 1).
where Cio and Ci0 are the concentrations of a reference immobile species (i) in the precursor (in this case either Ti or Al) and the alteration selvedge, respectively. Figure 6 presents concentration ratios for alteration selvedge v. precursor for series 28; the most altered sample is represented by sample 28/6 (located at 0–10 mm from the pegmatite margin). To minimize potential primary lithological compositional variation, possible over large sampling distances, we chose sample 28/5 (800 mm from pegmatite margin), rather than sample 27 or 30 (20 000 mm and 1100 mm, respectively), as the relatively unaltered reference precursor with which to judge relative elemental enrichment or depletion in the highly metasomatized sample (28/6). Figure 6a indicates marked mass addition of Na, Rb, P, Th, U, Pb, the REE and Y. All REE except Eu underwent progressive enrichment with alteration. The heavy REE (HREE; Gd to Lu) and Y underwent preferential enrichment relative to the light REE (LREE; La to Sm) across the traverse, with La/Lu values decreasing from 360 in sample 28/5 to 236 in the most altered sample (28/6; Fig. 6b). A notable feature of the REE patterns is that Eu underwent negligible mass change during alteration. This Eu ‘anomaly’ is further illustrated in Figure 6c, which demonstrates increasing values of Sm/Eu with increasing alteration, highlighting progressive enrichment of Sm over Eu. Si, Mg and Zr do not show significant mass addition or depletion (,+5%). In order of increasing mass depletion, Ca, Sr, Ba and K were lost from the rock (Fig. 6a). Unlike K, Rb underwent marked mass enrichment up to c. 80%. K/Rb ratios decreased markedly commencing c. 200 mm from the pegmatite (Table 9). Total mass change within the alteration zone, however, was negligible, no greater than c. 24%, equivalent to the loss of 4 g for every 100 g of rock (Table 11). Density differences between unaltered and highly altered samples (Table 9) are negligible. Therefore, the volume change that resulted from fluid–rock interaction was very small. The most noticeable feature of the concentration ratio diagram for sample 76 (Fig. 7) is the spectacular enrichment of K and Rb (c. 1100% and c. 4000%, respectively!). In contrast to sample 28, K and Rb display sympathetic behaviour. K/Rb values decrease significantly toward the pegmatite. Although some scatter in K/Rb is evident for more distal samples (Table 10) this presumably reflects low abundance of these components in the less altered material. The masses of P and Y also increase markedly, between c. 200 and 250%. Significant mass depletion of Na (c. 70%) is evident, whereas Ca (and Sr) and Fe may have undergone modest mass changes, between 10 and 20% depletion and 10 and 24% enrichment, respectively.
METASOMATISM OF THE NAPIER COMPLEX
303
Thorium, U and Pb underwent marked enrichment during alteration (Fig. 7). Total mass change for series 76 is zero within error (Table 11); estimates range between c. 23% (Ti reference frame) and c. þ9% (Al). Density increases from 2.95 g cm23 for precursor (e.g. sample 76/8B) to 3.12 g cm23 for the most altered selvedge (sample 76/10B), primarily as a result of the appearance and modal increase of garnet toward the pegmatite. The changes in the ratios K/Na, K/Ca and Ca/ Na are illustrated in Figure 8a and b with respect to increasing alteration in both traverse profiles. In traverse series 28, K/Na and Ca/Na decrease with increasing alteration as the pegmatite is approached (Fig. 8a). K/Ca decreases toward the pegmatite, then increases slightly immediately adjacent to the pegmatite margin. The length scale of alteration for K, Ca and Na was c. 400–300 mm from the pegmatite margin. For traverse sample series 76, in contrast, K/Na and K/Ca increase strongly near the contact with the pegmatite, and significant changes commence at c. 80 mm from the pegmatite margin (Fig. 8b).
Fig. 7. Concentration ratio diagram for series 76B. Annotation as for Figure 6a. The vertical scale break in the vertical axis to account for the mass increase of both K and Rb should be noted; also, in particular, the open circle for calculated Ca based on modes, mineral compositions and molar volumes of major phases. The calculated concentration ratio for Ca lies above the zero mass gain or loss reference line (Ca mass gain), whereas the concentration ratio for Ca determined from whole-rock geochemistry lies below the reference line, indicating Ca mass loss.
Mineral chemistry Feldspar. Single plagioclase and alkali feldspar crystals do not reveal any significant compositional zoning in series 28 (Table 3). Furthermore, within single samples, the composition of the exsolved plagioclase and alkali feldspar component of mesoperthite is the same as that of single matrix grains of plagioclase and alkali feldspar. There is, however, some minor compositional variation in the feldspars with increasing alteration toward the pegmatite. Alkali feldspar (and exsolved alkali
Little Si mass change is evident. The REE underwent considerable enrichment with progressive alteration in series 76, particularly the middle REE (MREE). A notable expection is Eu, which underwent far less mass enrichment relative to adjacent species.
Table 11. Mass changes for major components in series 28 and 76 Series 28 v. protolith
SiO2 Fe2O3 MgO CaO Na2O K2O P2O5 Ba Rb Sr Y P REE Total mass D
Series 76 v. protolith
% massTiO2
% massAl2O3
% massTiO2
% massAl2O3
24.0 6.8 216.2 218.8 15.3 228.0 19.5 224.3 69.5 222.9 69.1 31.7
0.8 12.0 212.0 214.7 21.1 224.4 25.4 220.5 78.0 219.1 77.5 41.5
210.9 10.5 1.0 221.5 272.1 1109.4 206.3 78.5 4035 269.1 222.6 264.2
0.3 24.3 13.6 211.7 268.6 1260.4 244.5 100.7 4551 265.2 262.9 309.7
24.4
20.4
23.4
8.7
Mass changes calculated for most altered sample (near pegmatite) relative to protolith sample 28/5 (series 28) and 8B (series 76).
304
C. J. CARSON & J. J. AGUE
are low at c. 0.03 wt%: FeO is low and variable, ranging between c. 0.10 and 0.25 wt%. Pegmatite 29 and 76 feldspar compositions and end-member activities are listed in Table 3. Kfeldspar from pegmatite 29 is Or85Ab14An,1; K-feldspar hosts thin exsolution lamellae (c. 5 mm wide) of almost pure albite (Or,1Ab95An5). Reintegration of this exsolved albite component (83.1 mol% Kfeldspar:16.6 mol% albite exsolution; Table 3) into the K-feldspar, gives a net composition of the pegmatite K-feldspar of Or70Ab28An1. We infer this to approximate the composition of the pegmatite Kfeldspar prior to exsolution during post-emplacement cooling. The plagioclase composition for pegmatite 29 is Or1Ab87An11. K-feldspar in pegmatite 76 is Or86Ab14, but also contains 10.5 mol% pure albite exsolution lamellae (Ab100). Reintegration yields a bulk, pre-exsolution pegmatite K-feldspar composition of Or77Ab23. Plagioclase is Or2Ab88An10.
Fig. 8. Selected ratio plots for traverses across (a) series 28 and (b) series 76B. The horizontal axis is distance from pegmatite margin (mm). The vertical dashed line in (a) represents the approximate location of the orthopyroxeneout isograd, which coincides with the location of the visible outer boundary of the alteration zone, clearly visible in Figure 2.
feldspar component of mesoperthite) from the precursor (sample 27) is Or88 – 90Ab9 – 11; alkali feldspar in the most altered sample (28/6) is slightly more potassic and less sodic (Or92Ab6 – 7). Barium contents for alkali feldspar are variable but generally 0.5 –0.8 wt% BaO, and show no discernible correlation with degree of alteration. Within the precursor, plagioclase is Or1Ab69An30; in contrast, plagioclase from the most altered sample 28/6 is markedly more sodic (Ab77An23). BaO in plagioclase is negligible (,0.1 wt%). No K-feldspar is present in series 76. Plagioclase has minimal core–rim compositional variation with single samples, and minor variation between samples. The composition of plagioclase immediately adjacent to the pegmatite is Or,1Ab65 – 71An28 – 32 (sample 76/2C). Plagioclase distal from the pegmatite is marginally more calcic and potassic and less sodic (Or1Ab62An37; sample 76/10C). BaO contents
Pyroxene. Orthopyroxene (Table 4) from series 28 has low Al2O3 content (c. 1.0 wt%), and enstatite (XEn) and wollastonite (XWo) mole fractions of 0.49 –0.50 and 0.01, respectively. Clinopyroxene has marginally higher Al2O3 contents (1.7–1.9 wt%) with XEn c. 0.34 and XWo c. 0.46. Orthopyroxene from series 76 also contains low Al2O3 (c. 1.2 wt%), and has XEn of c. 0.41 and XWo c. 0.02. No clinopyroxene is present in series 76. No compositional zoning was detected in analysed pyroxenes from sample series 28 or 76. Garnet. Garnet (Table 5) in sample 76 is almandine–pyrope–grossular. Minor compositional variation was found both as zoning within grains or between grains and samples. Typically, almandine mole fractions (XAlm) are in the range of c. 0.69– 0.73; XPy 0.13–15; XGr 0.10–14, and XSpss values are low at c. 0.02. Garnets in pegmatite 76 are almandine–spessartine (XSpss 0.18, XAlm c. 0.71) having low grossular and pyrope mole fractions of c. 0.04 and c. 0.06, respectively. Garnet chemistry for pegmatite 29 has been previously described (Carson et al. 2002b). Apatite group. Apatite from series 28 and 76 are fluorapatites (typically 2.5 wt% F) with a minor chlorapatite component (Table 6). Apatite from the most altered sample within the series 28 selvedge (28/6) has Ce2O3, La2O3 and Pr2O3 contents that are low and variable (0.04–0.06 wt%) and Y2O3 contents of 0.16 wt%. Apatite grains from the precursor are similar, although they contain slighter higher amounts of Ce2O3 and Nd2O3 (0.18 wt%) and Y2O3 (0.30 wt%). Minor SiO2 is may also be present (0.2 wt %) in apatite, substituting for P. Apatite grains from series 76 are similar to those from series 28.
METASOMATISM OF THE NAPIER COMPLEX
Clinozoisite–epidote–allanite group. Clinozoisite – epidote group minerals exhibit considerable variation in chemistry in series 28, particularly in the extent of REE substitution (Table 6). Clinozoisite–epidote that occurs as single, twinned, subhedral crystals associated with biotite clusters does not contain elevated REE or Y (e.g. Fig. 4d; Table 6), and contains c. 79% pistacite end-member (Fe3þ c. 0.79 p.f.u.: 12.5 O, assuming all Fe as Fe3þ for clinozoisite). In contrast, clinozoisite –epidote overgrowths on apatite (Fig. 4d) are characterized by highly variable allanite substitution (REE and Y replacing Ca on the A site). Measured Ce2O3, La2O3 and Nd2O3 are as much as 11.6, 6.4 and 2.4 wt%, respectively (Table 6). Assuming the coupled substitution Ca2þ þ Fe3þ ¼ REE3þ þ Fe2þ, Fe3þ was converted to Fe2þ by an amount equivalent to the amount of REE p.f.u. Magnesium contents are low and variable (0.03 wt% MgO); however, the REE-enriched varieties have slightly elevated MgO (0.3–0.4 wt%). Clinozoisite– epidote in series 76 is also typically enriched in the allanite end-member with c. 7 wt% Ce2O3, c. 3 wt% La2O3, and c. 4 wt% Nd2O3. Biotite. No significant compositional variation either within or between samples was observed in biotite. For series 28, biotite typically contains moderately elevated Ti (4.0–4.5 wt% TiO2) and has XFe c. 0.65 (Table 7). Cl contents are low at 0.02–0.04 wt% and F contents are c. 0.2 wt%. In comparison, series 76 biotites contain less TiO2 (3.0–3.5 wt%), XFe is 0.45–0.47, and Cl and F contents are c. 0.02 wt% and c. 0.45 wt%, respectively. Biotite from pegmatite 76 is more Fe rich (XFe ¼ 0.65), has TiO2 contents c. 0.5 wt%, and has Cl and F contents c. 0.08 and 0.07 wt%, respectively. Biotite compositions from pegmatite 29 have been previously published (Carson et al. 2002b). Amphibole group. Hornblende in series 28 (Table 7) displays some compositional variation, although most are calcic ferro-tschermakitic hornblendes. In series 28, matrix hornblende contains between 12.3 and 13.6 wt% Al2O3, c. 1 wt% TiO2 and XFe is c. 0.64. Na and K contents are low (K þ Na ,0.6 per 23 O); Cl contents are low (c. 0.02 wt%) and F was variable ranging from c. 0.22 wt% to more typical values of c. 0.08 wt%. In sample 28/5, hornblende can occur as residual moats, interpreted as outlining consumed orthopyroxene. These hornblendes are less aluminous (c. 8 wt% Al2O3), and have lower XFe (c. 0.52) and TiO2 contents (c. 0.4 wt%). Hornblende in series 76 has lower XFe (0.53–0.57) than matrix hornblende from series 28 and negligible compositional variation was observed across the sampling traverse (76/10 to 76/2). Grunerite– cummingtonite (series 76, Table 7) reveals little compositional variation and contains minor amounts of
305
Al2O3, CaO (both ,1 wt%) and TiO2 (c. 0.2 wt%). Both Na and K are low with ,0.03 cations per 23 oxygens. XFe is intermediate at 0.47–0.50.
Th – Pb monazite geochronology Sixteen analyses conducted on 10 monazite grains from pegmatite 76 (Fig. 3) yielded a weighted mean 208 Pb/232Th age of 532 + 8 Ma (n ¼ 18; Table 8) with excess scatter beyond that expected for analytical error alone (MSWD c. 10). This age is somewhat older than obtained by U–Pb and Th–Pb monazite analyses on pegmatite 29 by Carson et al. (2002b), which yielded weighted mean ages of 499.7 + 3.4 Ma (206Pb/238U), 496 + 2.9 Ma (207Pb/235U) and 498 + 1.7 Ma (208Pb/232Th). These results are broadly consistent with previous estimates of Early Palaeozoic pegmatite emplacement in the region (c. 520 Ma, Grew & Manton 1979; 522 + 10 Ma, Black et al. 1983a). From these monazite data, together with the isotopic data from previous studies, we conclude that emplacement of felsic post-tectonic pegmatites within the Napier Complex occurred during the interval 500– 530 Ma.
Whole-rock reactions To further assess K, Na and Ca mass exchange, in particular, during fluid –rock interaction that formed the alteration selvedges, the method of Brimhall (1979) and Ague (1994b) is utilized. This method identifies a suitable index mineral to establish an arbitrary reaction progress variable (j) to which the moles of other reactant and product phases can be referenced. We have defined the reaction progress variable (j) so that j is zero for the reference precursor (unaltered wall rock). If, during progressive alteration, the index mineral were to be fully consumed the reaction progress variable would go to unity. However, in the two examples studied here, the index mineral is not fully consumed, and therefore the reaction progress variable does not reach unity, even for the most altered sample immediately adjacent to the relevant pegmatite. Using the modal abundances (Tables 1 and 2) of the minerals across the traverses in both samples, known mineral molar volumes (see the Appendix) and measured rock density values (Tables 9 and 10), the moles of minerals per reference mass of rock participating in a reaction can be calculated for each sample. The ‘reference mass’ is determined by converting the amount of moles of a selected index phase per unit volume into moles per unit rock mass based on the measured rock density, then normalized to give 1 mole of that index phase within the reference precursor (i.e. mass*). Plotting the moles of minerals (per mass*) v. j yields lines whose slopes give the
306
C. J. CARSON & J. J. AGUE
Fig. 9. Reaction progress (j) diagrams for major phases, based on the technique of Brimhall (1979). Horizontal axis is extent of reaction indexed to K-feldspar for series 28 and plagioclase for series 76C. Vertical axis is moles of phases consumed or produced per reference mass of rock. Positive slope indicates production of phase; negative slope indicates consumption of phase. (a) Series 28; vertical dashed line indicates the approximate location of the pyroxeneout isograd, seen as the visible boundary of the alteration zone in Figure 2.
stoichiometric coefficients for the minerals in the alteration reactions. Given the stoichiometry for solids, fluid– rock reactions and mass balance can then be determined for fluid species of interest. The mass balance and rock density measurements demonstrate that volume changes and bulk mass changes attending alteration were negligible. Reaction progress (j) should not be confused with length scale or distance from the pegmatite; instead, it represents a quantitative indicator of mineralogical changes as a result of fluid –rock reaction. We have chosen K-feldspar as the index mineral for sample traverse 28 as it was abundant, easily identified in thin section after staining, and clearly decreases in modal abundance as the fluid–rock reaction proceeds. For sample traverse 76, we used plagioclase for similar reasons. Series 28. The results of this analysis for series 28 are illustrated in Figure 9a. The mass of the reference precursor containing 1 mole of K-feldspar is 1056 g (¼ mass*). The stoichiometric coefficient
for each mineral phase is readily determined from the slope of the line (Fig. 9a) across the reaction progress variable range (j from 0 ¼ unaltered precursor ! 0.6 most altered) with phase consumed (negative slope) or produced (positive slope). General inspection of Figure 9a illustrates that the alteration zone is characterized by the loss of clinopyroxene and orthopyroxene (at j ¼ 0.1) and this ‘pyroxene-out’ isograd is responsible for the development of the visible outer boundary of the alteration selvedge. Across the reaction progress range (0–0.6), plagioclase, biotite, apatite and epidote are produced and ilmenite, K-feldspar and quartz are consumed. Magnetite is not well defined but appears to generally increase in mode. Hornblende is a special case in that there was no net production or consumption, but as much as 0.05 mole was produced between j ¼ 0.1 and 0.4 and was consumed between j ¼ 0.4 and 0.6. From the slopes of the lines in Figure 9a it is straightforward to determine that for every mole of K-feldspar consumed, for example, 0.34 mole of
METASOMATISM OF THE NAPIER COMPLEX
307
Fig. 9. (b) Series 76C. For apatite: B, apatite moles/mass* determined from the modal calculations; W, apatite moles/mass* determined on P contents from the XRF results on series 76B (assuming apatite is the only significant P phase); the geochemical calculation of apatite moles/mass* is scaled horizontally to equate with the reaction coefficient range (j from 0 to 0.7) and is shown for comparison only. The numbers in parentheses for apatite are distance in millimetres from the pegmatite for the geochemical calculation of apatite moles/mass*.
plagioclase and 0.39 mole of biotite were produced. The net mass changes for the altered rock can be easily estimated by calculating the change in the total number of moles of a particular species across the reaction progress range, utilizing the measured mineral compositions (Tables 3–7), known molar volumes of participating phases (see the Appendix) and modal changes for each phase as a result of alteration (e.g. Brimhall 1979; Ague, 1994b; Fig. 9a and Tables 1 and 2). Such molar calculations across the reaction progress range of relevance here for series 28 (j between 0.0 and 0.6), indicate loss of c. 0.163 mole Kþ, c. 0.059 mole Ca2þ and gain of c. 0.178 mole Naþ across the reaction zone (per reference rock mass*). Presenting the calculated mass changes in terms of a concentration ratio (e.g equation (1); the ratio of final number of moles relative to the initial number of moles of a specific species) gives 1.18 (Naþ; 17.6% mass addition), 0.838 (Ca2þ; 16.2% mass loss) and 0.815 (Kþ; mass loss 18.5%). These calculations based on modal abundances provide remarkable independent agreement with the concentration ratios based on whole-rock geochemistry observed in series 28 (Fig. 6a and Table 11). The above calculations assume that plagioclase is the major Ca repository;
neglecting the Ca component in subordinate Ca-bearing phases, such as clinopyroxene, hornblende and apatite, has little overall influence on the results. Series 76. The mineral modal changes relative to reaction progress across series 76C are illustrated in Figure 9b. Plagioclase is the index mineral in this example and the mass of the reference precursor containing 1 mole of plagioclase is 639 g (¼ mass*). The reaction progress ranges from j ¼ 0 for least altered precursor to j ¼ 0.7 to most altered sample and plagioclase is not fully consumed (i.e. j does not go to unity). Mineral compositions are shown in Tables 3–7 and rock densities in Table 10. By inspection of Figure 9b, mineral modal changes series 76C is characterized by an overall modal increase of garnet, hornblende, biotite and apatite with progressive alteration. Plagioclase reduces in modal abundance, and although not well defined, quartz and grunerite also experience slight modal decreases. Orthopyroxene is completely consumed by j c. 0.1. Further inspection of Figure 9b shows a marked increase in production of biotite, hornblende and apatite from j ¼ 0.4 to 0.7. To better characterize the modal changes in
308
C. J. CARSON & J. J. AGUE
Figure 9b, we have subdivided biotite, hornblende and apatite into two reaction domains based on the observed rate of modal change relative to j. The first domain is from j ¼ 0.0 to 0.4 characterized by relatively little modal change in biotite, hornblende and apatite; the second domain is from j ¼ 0.4 to 0.7, where the modal change of these phases is markedly increased. It should be noted that, in contrast to series 28, which utilizes an essentially pristine precursor as a starting point allowing an accurate assessment of the overall fluid –rock reaction, pristine precursor was not, unfortunately, collected for series 76C. Although Figure 9b is illustrative for mineralogical development of the series 76 alteration selvedge, the original precursor probably contained the assemblage quartz, plagioclase and orthopyroxene as primary phases. The least altered sample in series 76C (76/10C) contains garnet and grunerite as pseudomorphs after orthopyroxene, indicating that grunerite and garnet are not peak UHT metamorphic phases. Grunerite is less abundant in the most altered sample (76/2C) than in the least altered sample (76/10C), giving the erroneous impression that it is consumed overall. However, this is an artefact of the abundance of grunerite in the least altered sample available. Grunerite is clearly an alteration product and produced by fluid–rock reaction. Similarly, garnet probably was not present in the precursor (based on textural evidence). This should be borne in mind when considering mineralogical and geochemical estimates of mass transfer of series 76C and as such these calculated estimates should be considered minimum values. Determined from the slopes of the regressions in Figure 9b, for every mole of plagioclase consumed, 0.23 mole of garnet is produced (Fig. 9b). Negligible biotite production occurred from j ¼ 0.0 to 0.4, whereas from j ¼ 0.4 to 0.7, biotite was produced at the rate of 0.95 mole per mole of plagioclase consumed. However, overall (from j ¼ 0.0 to 0.7) c. 0.26 mole of biotite was produced. Similarly, apatite production between j ¼ 0.0 and 0.4 was negligible, whereas 0.05 mole was produced between 0.4 and 0.7. Between j ¼ 0.0 and 0.4, 0.03 mole of hornblende was produced, and 0.08 mole was produced between j ¼ 0.4 and 0.7. Overall, 0.11 mole of hornblende was produced. Mass changes of Ca, Na and K accompanying alteration in series 76C are primarily controlled by the modal proportion and composition of the feldspars, hornblende and apatite (Ca) and biotite (K). Because of very low and/or irregular modal abundance (Table 2), epidote is not considered in mass balance determinations for Ca. Potassium, and Na mass balance, based on mineral modal changes and compositions, indicate gain of 0.21 mole of Kþ, and 0.40 mole Naþ across the reaction zone
(per mass*). In terms of concentration ratios these mass changes are 10 for Kþ (900% mass gain!) and 0.36 for Naþ (60% mass loss). As with series 28, these results provide excellent independent confirmation of the K and Na mass changes based on whole-rock geochemistry observed for series 76 presented above (Fig. 7). However, the calculated mass change for Ca in series 76C does not satisfactorily agree with the geochemical whole-rock data (Fig. 7). The mass change for Ca based on the observed mineral modal changes and measured compositions indicates 36% mass gain (concentration ratio of 1.36), whereas the mass change derived from the geochemical data is around 19% mass loss. One explanation for this discrepancy is that the mass change calculations for Ca will be strongly influenced by the observed apatite mode and that any imprecision in the measurement of the relatively low modal abundance of apatite will significantly affect the Ca mass change determinations. To assess this possibility, the mode of apatite (and the moles of apatite per mass*) was calculated from the whole-rock geochemistry (series 76B; Table 10), reasonably assuming that apatite was the only P-bearing phase in the rock. The results are illustrated in Figure 9b; although there is close agreement of apatite molal abundances in the comparatively unaltered samples (c. 0.01 mole of apatite per mass*) it is apparent that there is some divergence between measured apatite moles (based on modal observations) and the values determined from wholerock geochemistry (open circles, Fig. 9b). These observations suggest that measured apatite mode has been overestimated in the more altered samples across series 76C, contributing to an apparent Ca mass increase across the alteration zone. Substituting only the modal abundances of apatite calculated from whole-rock geochemistry for the observed apatite modes, recalculated mass changes for Ca return c. 12% mass gain (concentration ration c. 1.12; Fig. 7). Although this does not fully account for the Ca ‘discrepancy’, we presume that uncertainties in modal determination of other major Ca repositories (e.g. hornblende) may also contribute to the difference. We conclude that, for Ca in series 76, the geochemical determination, which indicates Ca mass loss, presented above (Fig. 7b; Table 10) is the preferred evaluation.
Discussion Contrasting patterns of K, Na and Ca metasomatism Sandiford (1985) concluded that there was no significant introduction of K or Rb in the development of retrograde schists in the Khmara Bay region
METASOMATISM OF THE NAPIER COMPLEX
(Fig. 1). In contrast, Sandiford (1985) also noted that related retrograde mylonite zones, associated with emplacement of significant volumes of pegmatite (probably Early Palaeozoic in age), are enriched in Si, K, Rb and Ba relative to the unaffected adjacent host lithology. The alteration selvedges adjacent to Early Palaeozoic pegmatites addressed here suggest contrasting patterns of K, Na and Ca geochemical patterns, which, we will show, are fundamentally controlled by the wall-rock feldspar mineralogy and composition. Figures 6 and 8a highlight the salient features of the whole-rock geochemical systematics for Na, Ca and K in the sample traverse series 28. Notably, there is a small progressive loss of K relative to Na with increasing wall-rock alteration. This is accompanied by a marked decrease in the modal abundance of K-feldspar (Table 1). Textural observations indicate increasingly ragged appearance of K-feldspar, particularly K-feldspar lamellae within mesoperthite, with increasing wall-rock alteration. These observations demonstrate that wall-rock Kfeldspar was unstable under the prevailing conditions during fluid influx. Similarly, though less well-defined, mass reduction of Ca contrasts with mass enrichment of Na, resulting in reduction of Ca/Na ratios with alteration. Minor modal increase of plagioclase is observed with progressive alteration (30–36%; Table 2), which is accompanied by recrystallization, and sub-grain development. Notably, plagioclase undergoes a compositional change from Ab70 in distal samples (e.g. 27 and 28/5; Table 3), becoming more albitic (Ab78) adjacent to the pegmatite (28/6). The compositional and modal changes account for the relative changes in Na and Ca in the wall-rock, as illustrated in Figure 6a. The modal and compositional changes of feldspars in series 28 dictate the geochemical patterns of K, Ca and Na within the wall-rock. The infiltrating fluid, in turn, fundamentally controlled the modification of wall-rock feldspars. As the alteration selvedge is spatially associated with pegmatite emplacement, the infiltrating fluid was presumably equilibrated with the pegmatite feldspar mineralogy. Alkali-feldspar from the wall-rock (series 28) is considerably more potassic (Or90, aOr 0.91) and less albitic (Ab6 – 10, aAb 0.27–0.42) than alkalifeldspar from the pegmatite 29 (Or70Ab28; aOr 0.79, aAb 0.81, following reintegration of exsolved albite components, Table 3). Consequently, the wall-rock alkali-feldspar was in disequilibrium with the invading, relatively K-poor fluid, resulting in decomposition of wall-rock alkali-feldspar and migration of K down a chemical potential gradient towards the pegmatite. Similarly, plagioclase in the precursor is markedly less albitic (Ab70) than the pegmatite plagioclase (Ab87; Table 3), which
309
resulted in progressive recrystallization of the precursor plagioclase to more albitic compositions nearer the pegmatite during fluid infiltration. In contrast to series 28, series 76 is strongly enriched in K in the highly altered wall-rock (Figs 7 and 8). The mafic host lithology contains no major primary mineral repositories for K. As a result, the chemical potential of K in the wall-rock was less than that of the invading fluid, resulting in diffusion of K through the pore fluid into the wall-rock, stabilizing biotite immediately adjacent to the pegmatite. In contrast, the net effect is mass loss of Na, and (if the whole-rock geochemistry reference is used) Ca. Mass change of Na and Ca was regulated by plagioclase modal abundance and, in the case of Ca, the growth of hornblende, garnet and apatite. Plagioclase in the wall-rock was less albitic (Ab63, aAb 0.57, aAn 0.57) than the pegmatite plagioclase (Ab88, aAb 0.88, aAn 0.12), resulting in compositional readjustment of the wall-rock plagioclase to more albitic compositions (to Ab71) during fluid influx. The compositional changes were accompanied by considerable modal decrease in wall-rock plagioclase. In a study of alkali metasomatism and the role of feldspars, Orville (1962) emphasized the relationship between fluid composition, temperature and coexisting feldspar compositions. Orville noted that a fluid in equilibrium with a two-feldspar assemblage will become poorer in K with decreasing temperature, thereby enriching K in wall-rocks through which such a fluid passes. A fluid in equilibrium with two feldspars will, if the fluid is flowing down a temperature gradient (down-T ), drive partial replacement of any pre-existing albite by Kfeldspar and raise the K/Na ratio of the local host rock, if local equilibrium with the fluid is to be maintained (e.g. Orville 1962; Dipple & Ferry 1992; Ague 1997). Conversely, a fluid flowing up-T will consume wall-rock K-feldspar, resulting in K loss, and enrich wall-rocks in Na, lowering the K/Na ratio. However, simple up- and down-T fluid flow is not the only mechanism by which alkali metasomatism might occur. Patterns of K, Na and Ca metasomatism are also strongly dependent on the composition of wall-rock feldspars that interact with any invading fluid, a point alluded to by Orville (1962, p. 308). For example, in series 28, we observe significant modal reduction of wall-rock K-feldspar with concomitant increase in the albite component of plagioclase. If the fluid accompanying pegmatite emplacement were assumed to have been in local equilibrium with the pegmatite two feldspar assemblages, then the invading fluid would have been undersaturated with K and Ca and oversaturated with Na, relative to wall-rock feldspar compositions (Table 3). The net effect, during fluid influx, was the breakdown
310
C. J. CARSON & J. J. AGUE
of the wall-rock K-feldspar, net mass loss of K from the wall-rock, and increase in the albite component of wall-rock plagioclase. Overall, K/Na is progressively reduced with increasing alteration (Fig. 8b). In series 76, the fluid in equilibrium with plagioclase (Ab88) in pegmatite 76 is oversaturated in Na and undersaturated in Ca relative to the wall-rock plagioclase in series 76 (Ab63), resulting in the increase of the albite component with increasing alteration of the wall-rock plagioclase (Ab71) adjacent to the pegmatite. However, overall Na was lost (Fig. 7) in series 76, as a result of significant modal reduction of plagioclase (c. 0.7 mole per mass*), the major repository of Na. Net loss of Ca from the wall-rock is controlled by the invading fluid being under-saturated in Ca and by the modal reduction of plagioclase. More extensive loss of Ca is presumably countered by the growth of hornblende and apatite (Fig. 9b) during progressive alteration. The examples we have described display contrasting K, Na and Ca ratios with increasing alteration. We emphasize that the observed contrasting K, Na and Ca ratios were strong functions of local rock disequilibrium with the infiltrating fluid. Infiltration of fluids that were out of equilibrium with wall-rocks will produce different patterns of K, Na and Ca enrichment and depletion, depending on the wall-rock feldspar compositions and the feldspar compositions with which the invading fluid is in equilibrium. As the feldspars in pegmatites 29 and 76 are similar in composition, it is reasonable to assume that the dissolved budget of the infiltrating fluid in both examples was also similar. However, as this fluid interacted with two wall-rock lithologies of fundamentally different feldspar assemblage and composition, the K, Na and Ca patterns produced in the wall-rock as a result of fluid interaction were also profoundly different. We note that caution should be employed when using wall-rock K, Na and Ca ratio patterns as a primary tool for determining fluid flow direction relative to thermal gradients. A final point that needs to be emphasized is that we consider that the low degree of partial melting expected at the H2O-saturated P–T conditions reported here (c. 8 kbar and c. 670 8C) did not significantly control the observed K, Ca and Na geochemical patterns. Extraction of a melt phase with a minimum melt composition would conceivably result in mass loss of K, Ca, Na, Si and Al relative to a relatively immobile or refractory reference component (e.g. Ti and/or Zr), a pattern that is not observed (Table 11). We conclude that if any partial melting occurred in the selvedges, the melt phase did not leave, probably because of the lack of penetrative deformation associated with pegmatite emplacement and the low volumes of melt present.
K/Rb patterns Sandiford (1985) noted that K/Rb ratios of amphibolite-facies schists and retrograde shear zones were lower than un-retrogressed Napier Complex granulites and thus eliminated the mantle as a possible fluid source. We too note a progressive reduction in K/Rb ratios across the series 28 and 76 traverses with increasing alteration (Tables 9 and 10), consistent with Sandiford’s findings. Interestingly, Rb behaves antithetically to K in series 28. This observation probably reflects the overwhelming influence of destruction of Kfeldspar and loss of K (c. 25% K mass loss). Although Rb (c. 75% mass gain) might be also expected to follow suit, any loss of Rb from breakdown of wall-rock alkali-feldspar was presumably counteracted by growth of alteration biotite, which grew in response to the infiltration of, and in equilibrium with, the low K/Rb fluid. In series 76, however, K and Rb are both strongly enriched (c. 1100% and c. 4000%, respectively; Fig. 7) in the wall-rock. The K/Rb ratio (Table 10) of the wallrock, which contains no primary K-bearing phases (Table 2), was strongly regulated by the growth of secondary biotite at or near equilibrium with the low K/Rb infiltrating fluid.
REE and Y patterns, mobility and relationship with P Historically, the REE have been considered nearly immobile during metamorphism, particularly during dry or fluid-absent conditions (e.g. Taylor & McLennan 1985). However, numerous studies now indicate that REE mobility can occur under a wide variety of geological conditions (e.g. Grauch 1989, and numerous references therein; Giere´ 1990; Pan & Fleet 1996; Ague 2003). Although a detailed treatment of the nature of REE mobility is beyond the purpose of this paper, several points regarding the behaviour of the REE in the examples studied here can be emphasized. As illustrated in Figures 6a and 7, relative to the reference precursor used, mass addition of the REE to the P wall-rock is evident, with enrichment c. REE for series 28 and 260–310% P 30–40% REE for series 76 (Table 11). In both cases REE enrichment is paralleled by mass addition of P2O5 (20 –25% for series 28; 200–250% for series 76; Table 11). Although several field-based studies have indicated the potential role of fluoride complexes facilitating REE transport during hydrous metamorphism and metasomatism (e.g. Giere´ 1990; Rubin et al. 1993; Bingen et al. 1996; Pan & Fleet 1996; Montero et al. 1998), in the examples studied here the fluid is of low –moderate halogen content (Carson et al. 2002b), and we
METASOMATISM OF THE NAPIER COMPLEX
conclude that F complexes are unlikely to have been significantly involved in REE transport. We note the correlation of P and Y enrichment with increasing alteration in both examples and suggest that phosphate and/or yttrium complexing within the infiltrating fluid may have played a significant role in transport of REE, a conclusion offered by other workers (e.g. Giere´ 1990; Ague 2003). A final point worth highlighting is that both alteration selvedges record a marked Eu ‘anomaly’, in that Eu displays little or no mass addition in contrast to other adjacent REE. We conclude that the Eu ‘anomaly’ indicates that the infiltrating fluid was Eu-depleted (relative to adjacent REE species). This strongly suggests that the fluids responsible for metasomatism of the wall-rocks were in equilibrium with crystallizing pegmatite plagioclase, depleting Eu available to the fluid.
Tectonic implications Many studies that examine the post-Archaean evolution of the Napier Complex referred to the contribution of Neoproterozoic (c. 1000 Ma) highgrade tectonism that evolved in the adjacent Rayner Complex (e.g. Grew et al. 1982; Sandiford & Wilson 1984; Harley 1985; Black et al. 1987). However, the potential significance of Early Palaeozoic events in the geological evolution of the Napier Complex will be examined in the following discussion. To account for post-Archaean rehydration of the ‘dry’ Napier Complex, an external source of fluid is required. Ellis (1983) proposed a number of potential sources, favouring the possibility of metamorphic dehydration of ‘younger wet sediments’ underplated beneath the Napier Complex, as a source of fluids. Similarly, Harley (1985) proposed that fluids could be released from the dehydration of underlying low-grade metamorphic lithologies. Ellis (1983) and Harley (1985) both concluded that underplating occurred during c. 1000 Ma tectonism. Sandiford (1985) argued that rehydration of the Napier Complex was partially accomplished by fluids introduced during emplacement of tourmaline- and beryl-saturated pegmatites, also derived from underplated hydrous lithologies, although the proposed timing of underplating was unclear. The spatial association of rehydration of the SW Napier Complex with the emplacement of peraluminous pegmatites of known early Palaeozoic age (e.g. Grew & Manton 1979; Black et al. 1983a) is evident and it would appear likely that Early Palaeozoic tectonism, at least in part, has played a significant role in the evolution of the Napier Complex. Sandiford (1985) noted that geochemical modification of Napier Complex gneisses during fluid influx and amphibolite-facies shear zone formation
311
is characterized by enrichment of large ion lithophile elements (elements typically depleted in high-grade terranes and mantle-derived fluids) and by the marked reduction of K/Rb ratios, suggesting a crustal source for fluids. Similarly, in this study, alteration selvedges adjacent to early Palaeozoic pegmatites are also variably enriched in a variety of components typically depleted in high-grade terranes (Pb, Th, U, Rb, P, Y and REE which may include K and Na) and are also characterized by lower K/Rb ratios. We reaffirm the suggestions of Ellis (1983), (1985) and Harley (1985) and Sandiford (1985) that the fluids, and accompanying pegmatites, that invaded the Napier Complex during the Early Palaeozoic were derived from dehydration and melting reactions affecting underplated hydrous low-grade or previously unmetamorphosed lithologies of crustal origin. However, we suggest that fluid influx into the Napier Complex may not have resulted primarily from Neoproterozoic (c. 1000 Ma) Rayner activity. This suggestion is based on two premises; first, the growing body of evidence of convergent Early Palaeozoic tectonism in adjacent regions and, second, the association of fluid influx with pegmatites of Early Palaeozoic age within the Napier Complex itself. Shiraishi et al. (1997) recently re-examined the chronological evolution of the Rayner Complex, immediately adjacent to the SW margin of the Napier Complex (at Mt. Vechernyaya and Forefinger Point). This region is also considered to represent the western limits of the c. 1000 Ma Rayner Complex (Black et al. 1987). However, Shiraishi et al. (1997) concluded that this region underwent high-temperature metamorphism during the Early Palaeozoic at c. 520–540 Ma and represents the eastward extension of the Early Palaeozoic granulite– amphibolite-facies Lu¨tzowHolm Complex. Furthermore, within the Napier Complex, Carson et al. (2002a, b) demonstrated that zircon in rehydrated Archaean gneisses adjacent to Early Palaeozoic (c. 500 Ma) pegmatites experienced isotopic disturbance with lower concordia intercepts that coincide with the timing of pegmatite emplacement and fluid influx. The commonly observed association of Early Palaeozoic pegmatites and the hydration of adjacent Archaean gneisses (e.g. Grew et al. 1982; Black et al. 1983a) in the Napier Complex is unequivocal. We therefore conclude that the fluids (and pegmatites) were derived from dehydration and partial melting of the aforementioned lithologies underplated beneath the Napier Complex during the Early Palaeozoic. The sporadic presence of beryl in the Early Palaeozoic pegmatites on Tonagh Island provides some additional tectonic constraints that are worth mentioning here. First, as cordierite is an important reservoir of Be (the Be partition coefficient for
312
C. J. CARSON & J. J. AGUE
cordierite –melt is as much as 207; e.g. London & Evensen 2002), the presence of beryl in felsic pegmatites suggests that cordierite was not present in the melt source region. Second, London & Evensen (2002) also argued that beryl saturation in pegmatites can be achieved only by the extraction of a highly fractionated residual melt of a larger magmatic body, which undergoes yet further fractionation to produce the required c. 70 ppm Be in the residual liquid necessary for beryl crystallization. These arguments suggest that partial melting in the underplated lithologies beneath the Napier Complex must have proceeded at pressures beyond cordierite stability, presumably in the garnet stability field, and that the observed pegmatites represent the highly fractionated end products of much larger magmatic systems at depth. The underplating of crustal sequences beneath the Napier Complex presumably evolved as a consequence of SW –NE crustal convergence during the early Palaeozoic, near-synchronous with tectonothermal activity in the adjacent Lu¨tzow-Holm
Complex (e.g. Shiraishi et al. 1997). It is also noteworthy that there is a general decrease in the volume of Early Palaeozoic pegmatites, from near the SW margin of the Napier Complex in the Casey Bay region (Grew 1981) to the Amundsen Bay region. Northeast of Amundsen Bay, early Palaeozoic pegmatites are rare or non-existent (E. S. Grew, pers. comm.). This obsesrvation also supports the general notion of an unmetamorphosed or low-grade crustal sequence being underthrust, dehydrated and metamorphosed beneath the SW Napier Complex during broadly NE– SW convergence during the Early Palaeozoic. Underplating of material beneath the Napier Complex is also consistent with the recent suggestion that Early Palaeozoic tectonism was a period of contraction, final amalgamation and assembly of the East Antarctic Shield (e.g. Boger et al. 2001, 2002). The salient structural features of the previous discussion are presented in a schematic interpretative diagram (Fig. 10) that is modified from Harley & Hensen (1990; their fig. 12.10).
Fig. 10. Schematic regional cross-section adapted and modified from Harley & Hensen (1990) illustrating possible relationships between the southwestern Napier Complex and the adjacent Early Palaeozoic Lu¨tzow-Holm Complex, based on the conclusions of Shiraishi et al. (1997). Section A– A0 shown in Figure 1 is shown. Molodezhnaya (Russia) and Syowa (Japan) are research stations. The Neoproterozoic Rayner Complex is shown, but is illustrated as reworked by Early Palaeozoic tectonism. Underplating of hydrous low-grade lithologies occurred during Early Palaeozoic crustal convergence, resulting in crustal thickening, uplift and, at least, partial exhumation of the overlying Napier Complex (as proposed by Harley 1985; Sandiford 1985; Harley & Hensen 1990). Significantly, prograde dehydration and partial anatexis of the fertile underplated crustal units resulted in subsequent influx of pegmatites and associated fluid into the overlying anhydrous Napier Complex during the Early Palaeozoic, resulting in the mass addition of exotic non-volatile elements. The location of the cordierite-out isograd is schematic to indicate that the region of partial melting is cordierite-absent, to partially account for beryl saturation of the pegmatites.
METASOMATISM OF THE NAPIER COMPLEX
Conclusions The emplacement of post-orogenic peraluminous pegmatites during the Early Palaeozoic into the Archaean Napier Complex was accompanied by the infiltration of aqueous, low-salinity fluids. Fluid infiltration resulted in rehydration, and facilitated recrystallization, of Archaean host orthogneisses at upper amphibolite-facies conditions (c. 8 kbar and c. 670 8C) immediately adjacent to pegmatites. Geochemical whole-rock data indicate that rehydration was accompanied by wall-rock enrichment of elements typically depleted in granulite-facies terranes and mantle sources; for example, U, Pb, Th, Rb and P (and which may include K and Na). This observation suggests that dehydration metamorphism of an underlying, initially unmetamorphosed or low-grade crustal lithology may have provided a source for aqueous fluids (a point emphasized by several previous workers) and peraluminous silicate melts into the Napier Complex. Underplating of previously unmetamorphosed lithologies underneath the Napier Complex during early Palaeozoic (500– 530 Ma) convergence tectonism is a plausible scenario. Metasomatism of orthogneiss precursors, as a result of fluid infiltration accompanying peraluminous pegmatite emplacement, resulted in contrasting behaviour of K, Na and Ca with progressive alteration. The fluid accompanying pegmatite emplacement was in equilibrium with pegmatite feldspars, but in local disequilibrium with wall-rock precursor orthogneiss feldspars. The modification of wall-rock feldspar compositions and the opensystem loss or addition of K, Na and Ca were fundamentally regulated by reactions driven by infiltration of pegmatite-derived fluids that were out of chemical equilibrium with the orthogneisses. This study highlights that the addition or loss of K, Na and Ca in metasomatized terranes need not be the result of equilibrated fluid flow along regional temperature or pressure gradients; infiltration of disequilibrium fluids can also provide a powerful mechanism for K, Na and Ca mass change. Geological material for this manuscript was obtained whilst CJC was a member of the 40th Japanese Antarctic Research Expeditions (JARE-40) during the austral summer of 1998 – 1999. C.J.C. extends thanks to K. Shiraishi and Y. Motoyoshi of the National Institute of Polar Research (NIPR) for invitation to participate in JARE-40. E. S. Grew kindly provided sample 76 (Grew sample number GS99012201) and the photograph in Figure 2. C.J.C. also thanks the expeditioners of JARE-39 and -40, particularly the Tonagh Island field party, and the officers and crew of the icebreaker Shirase for the warm hospitality and generosity during JARE-40. Assistance by, and discussion with, J. Eckert, Jr. during electron microprobe sessions at Yale University is appreciated. The manuscript was
313
improved by constructive reviews by J. Hollis and J. Hermann. Discussion with C. M. Breeding is also appreciated. C.J.C. acknowledges receipt of the Damon Wells ’58 fellowship at Yale University during the period from May 1999 to July 2001; analytical costs were supported by National Science Foundation grant EAR-0001084 and EAR-9810089.
Appendix The following molar volumes were used (cm3 mole21); biotite, 150.0 (O ¼ 12); quartz, 22.7 (O ¼ 2); plagioclase (An80) 100.0 (O ¼ 8); orthopyroxene, 64.3 (O ¼ 6); clinopyroxene, 66.2 (O ¼ 6); grunerite, 270.9 (O ¼ 22; hornblende (pargasite), 271.9 (O ¼ 23); epidote–clinozoisite, 136.8 (O ¼ 13); K-feldspar 108.9 (O ¼ 8), apatite 157.6 (O ¼ 13) and garnet 115.0 (O ¼ 12). These values are from Robie et al. (1978), Holland & Powell (1998) and Philpotts (1990); the molar volume is for the mineral formulae based on the indicated number of oxygens.
References A GUE , J. J. 1991. Evidence for major mass transfer and volume strain during regional metamorphism of pelites. Geology, 19, 855– 858. A GUE , J. J. 1994a. Mass transfer during Barrovian metamorphism of pelites, south–central Connecticut. I: Evidence for changes in composition and volume. American Journal of Science, 294, 989– 1057. A GUE , J. J. 1994b. Mass transfer during Barrovian metamorphism of pelites, south–central Connecticut. II: Channelized fluid flow and the growth of staurolite and kyanite. American Journal of Science, 294, 1061– 1134. A GUE , J. J. 1997. Crustal mass transfer and index mineral growth in Barrow’s garnet zone, north-east Scotland. Geology, 25, 73– 76. A GUE , J. J. 1998. Simple models of coupled fluid infiltration and redox reactions in the crust. Contributions to Mineralogy and Petrology, 132, 180–197. A GUE , J. J. 2003. Fluid infiltration and transport of major, minor, and trace elements during regional metamorphism of carbonate rocks, Wepawaug Schist, Connecticut, USA. American Journal of Science, 303, 753–816. A GUE , J. J. & R YE , D. M. 1999. Simple models of CO2 release from metacarbonates with implications for interpretation of directions and magnitudes of fluid flow in the deep crust. Journal of Petrology, 40, 1443– 1462. A GUE , J. J. & VAN H AREN , J. L. M. 1996. Assessing metasomatic mass and volume changes using the Bootstrap, with application to deep crustal hydrothermal alteration of marble. Economic Geology, 91, 1169– 1182. A YERS , J. C., M ILLER , C. F., G ORISCH , B. & M ILLEMAN , J. 1999. Textural development of monazite during high-grade metamorphism: hydrothermal growth kinetics, with implications for U–Th– Pb geochronology. American Mineralogist, 84, 1766–1780.
314
C. J. CARSON & J. J. AGUE
B EACH , A. 1976. The interrelations of fluid transport, deformation, geochemistry and heat flow in early Proterozoic shear zones in the Lewisian complex. Philosophical Transactions of the Royal Society of London, Series A, 280, 569– 604. B EACH , A. 1980. Retrogressive metamorphic processes in shear zones with special reference to the Lewisian Complex. Journal of Structural Geology, 2, 257–263. B ICKLE , M. J. & M C K ENZIE , D. 1987. The transport of heat and matter by fluids during metamorphism. Contributions to Mineralogy and Petrology, 95, 384– 392. B INGEN , B., D E M AIFFE , D. & H ERTOGEN , J. 1996. Redistribution of rare earth elements, thorium and uranium over accessory minerals in the course of amphibolite to granulite facies metamorphism: the role of apatite and monazite in orthogneisses from southwestern Norway. Geochimica et Cosmochimica Acta, 60, 1341–1354. B LACK , L. P. 1988. Isotopic resetting of U –Pb zircon and Rb– Sr and Sm– Nd whole-rock systems in Enderby Land, Antarctica: Implications for the interpretation of isotopic data from polymetamorphic and multiply deformed terrains. Precambrian Research, 38, 355– 365. B LACK , L. P. & J AMES , P. R. 1983. Geological history of the Archaean Napier Complex in Enderby Land. In: O LIVER , R. L., J AMES , P. R. & J AGO , J. B. (eds) Antarctic Earth Science. Cambridge University Press, Cambridge, 11–15. B LACK , L. P., J AMES , P. R. & H ARLEY , S. L. 1983a. Geochronology and geological evolution of metamorphic rocks in the Field Islands area, East Antarctica. Journal of Metamorphic Geology, 1, 277–303. B LACK , L. P., J AMES , P. R. & H ARLEY , S. L. 1983b. The geochronology, structure and metamorphism of early Archaean rocks at Fyfe Hills, Enderby Lands, Antarctica. Precambrian Research, 21, 197– 222. B LACK , L. P., F ITZGERALD , J. D. & H ARLEY , S. L. 1984. Pb isotopic composition, colour, and microstructure of monazites from a polymetamorphic rock in Antarctica. Contributions to Mineralogy and Petrology, 85, 141– 148. B LACK , L. P., S HERATON , J. W. & J AMES , P. R. 1986a. Late Archaean granites of the Napier Complex, Enderby Land, Antarctica: A comparison of Rb– Sr, Sm–Nd and U– Pb isotopic systematics in a complex terrain. Precambrian Research, 32, 343 –368. B LACK , L. P., W ILLIAMS , I. S. & C OMPSTON , W. 1986b. Four zircon ages from one rock: the history of a 3930 Ma old granulite from Mt Sones, Enderby Land, Antarctica. Contributions to Mineralogy and Petrology, 94, 427–437. B LACK , L. P., H ARLEY , S. L., S UN , S. S. & M C C ULLOCH , M. T. 1987. The Rayner Complex of east Antarctica, complex isotopic systematics within a Proterozoic mobile belt. Journal of Metamorphic Geology, 5, 1–26. B OGER , S. D., W ILSON , C. J. L. & F ANNING , C. M. 2001. Early Palaeozoic tectonism within the East Antarctic Craton: the final suture between East and West Gondwana? Geology, 29, 463–466. B OGER , S. D., C ARSON , C. J., F ANNING , C. M., H ERGT , J. M., W ILSON , C. J. L. & W OODHEAD , J. D. 2002. Pan-African intraplate deformation in the northern
Prince Charles Mountains, east Antarctica. Earth and Planetary Science Letters, 195, 195–210. B REEDING , C. M. & A GUE , J. J. 2002. Slab-derived fluids and quartz-vein formation in an accretionary prism, Otago Schist, New Zealand. Geology, 30, 499– 502. B RIMHALL , G. H., J R . 1979. Lithologic determinations of mass transfer mechanisms of multiple-stage porphyry copper mineralization at Butte, Montana: vein formation by hypogene leaching and enrichment of potassium silicate protore. Economic Geology, 74, 556–589. C ARSON , C. J., C LARKE , G. L. & P OWELL , R. 2000. Hydration of eclogite from the Pam Peninsula, New Caledonia. Journal of Metamorphic Geology, 18, 79–90. C ARSON , C. J., A GUE , J. J. & C OATH , C. D. 2002a. U –Pb geochronology from Tonagh Island, East Antarctica: implications for the timing of ultra-high temperature metamorphism of the Napier Complex. Precambrian Research, 116, 237– 263. C ARSON , C. J., A GUE , J. J., G ROVE , M., C OATH , C. D. & H ARRISON , T. M. 2002b. U –Pb isotopic behaviour of zircon during upper-amphibolite facies fluid infiltration in the Napier Complex, east Antarctica. Earth and Planetary Science Letters, 199, 287– 310. C HAMBERLAIN , C. P. & R UMBLE , D. 1988. Thermal anomalies in a regional metamorphic terrane; an isotopic study of the role of fluids. Journal of Petrology, 29, 1215–1232. D ALLWITZ , W. B. 1968. Coexisting sapphirine and quartz in granulite from Enderby Land, Antarctica. Nature, 219, 476– 477. D E P AOLO , D. J., M ANTON , W. I., G REW , E. S. & H ALPERN , M. 1982. Sm–Nd, Rb–Sr and U–Th–Pb systematics of granulite facies rocks from the Fyfe Hills, Enderby Land, Antarctica. Nature, 298, 614–618. D IPPLE , G. M. & F ERRY , J. M. 1992. Metasomatism and fluid flow in ductile fault zones. Contributions to Mineralogy and Petrology, 112, 149– 164. E LLIS , D. J. 1980. Osumilite–sapphirine–quartz granulites from Enderby Land, Antarctica: P –T conditions of metamorphism, implications for garnet– cordierite equilibria and the evolution of the deep crust. Contributions to Mineralogy and Petrology, 74, 201– 210. E LLIS , D. J. 1983. The Napier and Rayner Complex of Enderby Land, Antarctica—contrasting styles of metamorphism and tectonism. In: O LIVER , R. L., J AMES , P. R. & J AGO , J. B. (eds) Antarctic Earth Science. Cambridge University Press, Cambridge, 20– 23. E LLIS , D. J. & G REEN , D. H. 1985. Garnet forming reactions in mafic granulites from Enderby Land, Antarctica—implications for geothermometry and geobarometry. Journal of Petrology, 26, 633– 662. E LLIS , D. J., S HERATON , J. W., E NGLAND , R. N. & D ALLWITZ , W. B. 1980. Osumilite– sapphirine– quartz granulites from Enderby Land, Antarctica— Mineral assemblages and reactions. Contributions to Mineralogy and Petrology, 72, 123–143. F ERRY , J. M. 1996. Three novel isograds in metamorphosed siliceous dolomites from the Ballachulish aureole, Scotland. American Mineralogist, 81, 485–494. F ERRY , J. M. & D IPPLE , G. M. 1991. Fluid flow, mineral reactions, and metasomatism. Geology, 19, 211– 214. F ERRY , J. M. & G ERDES , M. L. 1998. Chemical reactive fluid flow during metamorphism. Annual Review of Earth and Planetary Sciences, 26, 255–287.
METASOMATISM OF THE NAPIER COMPLEX F UHRMAN , M. L. & L INDSLEY , D. H. 1988. Ternaryfeldspar modelling and thermometry. American Mineralogist, 73, 210 –215. G IERE´ , R. 1990. Hydrothermal mobility of Ti, Zr and REE: examples from the Bergell and Adamello contact aureoles (Italy). Terra Nova, 2, 60– 67. G RAUCH , R. I. 1989. Rare Earth Elements in metamorphic rocks. In: L IPIN , B. R. & M C K AY , G. A. (eds) Geochemistry and Mineralogy of Rare Earth Elements. Mineralogical Society of America, Reviews in Mineralogy, 21, 147–167. G REW , E. S. 1978. Precambrian basement at Molodezhnaya Station, East Antarctica. Geological Society of America Bulletin, 89, 801–813. G REW , E. S. 1980. Sapphirine þ quartz association from Archean rocks in Enderby Land, Antarctica. American Mineralogist, 65, 821 –836. G REW , E. S. 1981. Surinamite, taaffeite, and beryllian sapphirine from pegmatites in granulite facies rocks of Casey Bay, Enderby Land, Antarctica. American Mineralogist, 66, 1022–1033. G REW , E. S. 1982. Osumilite in the sapphirine– quartz terrane of Enderby Land, Antarctica; implications for osumilite petrogenesis in the granulite facies. American Mineralogist, 67, 762–787. G REW , E. S. 1998. Boron and beryllium minerals in granulite-facies pegmatites and implications of beryllium pegmatites for the origin and evolution of the Archean Napier Complex of East Antarctica. Memoirs of the National Institute of Polar Research, Special Issue, 53, 74– 92. G REW , E. S. & M ANTON , W. I. 1979. Archean rocks in Antarctica: 2.5 billion year uranium–lead ages of pegmatites in Enderby Land. Science, 206, 443 –444. G REW , E. S., M ANTON , W. I. & S ANDIFORD , M. 1982. Geochronological studies in East Antarctica: age of pegmatites in Casey Bay, Enderby Land. Antarctic Journal of the United States, 17, 1 –2. H ARLEY , S. L. 1985. Paragenetic and mineral-chemical relationships in orthoamphibole-bearing gneisses from Enderby Land, east Antarctica: a record of Proterozoic uplift. Journal of Metamorphic Geology, 3, 179–200. H ARLEY , S. L. 1987. Pyroxene-bearing granulites from Tonagh Island, Enderby Land, Antarctica: further evidence for very high temperatures (.950 8C) Archaean regional metamorphism in the Napier Complex. Journal of Metamorphic Geology, 5, 241–356. H ARLEY , S. L. 2003. Archaean–Cambrian crustal development of East Antarctica: metamorphic characteristics and tectonic implications. In: Y OSHIDA , M., W INDLEY , B. F. & D ASGUPTA , S. (eds) Proterozoic East Gondwana; Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 203– 230. H ARLEY , S. L. & B LACK , L. P. 1987. The Archaean geological evolution of Enderby Land, Antarctica. In: P ARK , R. G. & T ARNEY , J. (eds) Evolution of the Lewisian and Comparable Precambrian High-grade Terrains. Geological Society, London, Special Publications, 27, 285–296. H ARLEY , S. L. & B LACK , L. P. 1997. A revised Archaean chronology for the Napier Complex, Enderby Land, from SHRIMP ion-microprobe studies. Antarctic Science, 9, 74–91.
315
H ARLEY , S. L. & H ENSEN , B. J. 1990. Archaean and Proterozoic high-grade terranes of East Antarctica (408E– 608E): A case study of the diversity in granulite facies metamorphism. In: A SHWORTH , J. R. & B ROWN , M. (eds) High Temperature Metamorphism and Crustal Anatexis. Unwin Hyman, London, 320– 370. H ARLEY , S. L. & M OTOYOSHI , Y. 2000. Al zoning in orthopyroxene in a sapphirine quartzite: evidence for .1120 8C UHT metamorphism in the Napier Complex, Antarctica, and implications for the entropy of sapphirine. Contributions to Mineralogy and Petrology, 138, 293– 307. H ARRISON , T. M., M C K EEGAN , K. D. & L E F ORT , P. 1995. Detection of inherited monazite in the Manaslu leucogranite by 208Pb/232Pb ion microprobe dating: crystallisation age and tectonic implications. Earth and Planetary Science Letters, 133, 271– 282. H ARRISON , T. M., G ROVE , M., M C K EEGAN , K. D., C OATH , C. D., L OVERA , O. M. & L E F ORT , P. 1999. Origin and emplacement of the Manaslu intrusive complex, Central Himalaya. Journal of Petrology, 40, 3 –19. H OKADA , T. 2001. Feldspar thermometry in ultrahightemperature metamorphic rocks: evidence of crustal metamorphism attaining 1100 8C in the Archaean Napier Complex, East Antarctica. American Mineralogist, 86, 932–938. H OLLAND , T. J. B. & B LUNDY , J. 1994. Non-ideal interactions in calcic amphiboles and their bearing on amphibole –plagioclase thermometry. Contributions to Mineralogy and Petrology, 116, 433–447. H OLLAND , T. J. B. & P OWELL , R. 1998. An internallyconsistent thermodynamic data set for phases of petrological interest. Journal of Metamorphic Geology, 16, 309– 343. J AMTVEIT , B., B UCHER -N URMINEN , K. & A USTRHEIM , H. 1990. Fluid controlled eclogitization of granulites in deep crustal shear zones, Bergen arcs, Western Norway. Contributions to Mineralogy and Petrology, 104, 184–193. K ELLY , N. M. & H ARLEY , S. L. 2005. An integrated microtextural and chemical approach to zircon geochronology: refining the Archaean history of the Napier Complex, east Antarctica. Contributions to Mineralogy and Petrology, 149, 57– 84. K LAPER , E. V. 1990. Reaction-enhanced formation of eclogite-facies shear zones in granulite-facies anorthosites. In: K NIPE , R. J. & R UTTER , E. H. (eds) Deformation Mechanisms, Rheology and Tectonics. Geological Society, London, Special Publications, 54, 167– 173. L ONDON , D. & E VENSEN , J. M. 2002. Beryllium in silicic magmas and the origin of beryl-bearing pegmatites. In: G REW , E. S. (ed.) Beryllium; Mineralogy, Petrology and Geochemistry. Mineralogical Society of America, Reviews in Mineralogy and Geochemistry, 54, 445– 486. L UDWIG , K. R. 1999. User’s manual for Isoplot/Ex, v2.3, a geochronological toolkit for Microsoft Excel. Berkeley Geochronological Centre Special Publications, 1a. M ANNING , C. E. 1994. The solubility of quartz in H2O in the lower crust and upper mantle. Geochimica et Cosmochimica Acta, 58, 4831–4839.
316
C. J. CARSON & J. J. AGUE
M IYAMATO , T., G REW , E. S., S HERATON , J. W. ET AL . 2000. Lamproite dykes in the Napier Complex at Tonagh Island, Enderby Land, East Antarctica. Polar Geoscience, 13, 41– 59. M ONTERO , P., F LOOR , P. & C ORRETGE´ , G. 1998. The accumulation of rare-earth and high field strength elements in peralkaline granitic rocks: the Galin˜eiro orthogneiss complex, northwestern Spain. Canadian Mineralogist, 36, 683– 700. O RVILLE , P. M. 1962. Alkali metasomatism and the feldspars. Norsk Geologisk Tidsskrift, 42, 283–316. O SANAI , Y., T OYOSHIMA , T., O WADA , M., T SUNOGAE , T., H OKADA , T. & C ROWE , W. A. 1999. Geology of ultrahigh-temperature metamorphic rocks from Tonagh Island in the Napier Complex, East Antarctica. Polar Geoscience, 12, 1– 28. O SANAI , Y., T OYOSHIMA , T., O WADA , M. ET AL . 2001. Explanatory text of Geological Map of Tonagh Island, Enderby Land, Antarctica. Antarctic Geological Map Series (1 : 10 000), Sheet 38. National Institute of Polar Research, Tokyo. O WADA , M., O SANAI , Y. & K AGAMI , H. 1994. Isotopic equilibration age of Sm– Nd whole-rock system in the Napier Complex (Tonagh Island), East Antarctica. Proceedings of NIPR Symposium of Antarctic Geoscience, 7, 122– 132. P AN , Y. & F LEET , M. E. 1996. Rare earth element mobility during prograde granulite facies metamorphism: significance of fluorine. Contributions to Mineralogy and Petrology, 123, 251 –262. P ASSCHIER , C. W. 1985. Water-deficient mylonite zones—an example from the Pyrenees. Lithos, 18, 115– 127. P HILPOTTS , A. R. 1990. Principles of Igneous and Metamorphic Petrology. Prentice Hall, Englewood Cliffs, NJ. P OWELL , R. & H OLLAND , T. J. B. 1994. Optimal geothermometry and geobarometry. American Mineralogist, 79, 120 –133. R AVICH , M. G. 1972. Regional metamorphism of the Antarctic Platform crystalline basement. In: A DIE , R. J. (ed.) Antarctic Geology and Geophysics. International Union of Geological Sciences, Series B, 505– 515. R OBIE , R. A., H EMINGWAY , B. S. & F ISHER , J. R. 1978. Thermodynamic properties of minerals and related substances at 298.15 K and 1 Bar (105 Pascals) pressure and at higher temperatures. US Geological Survey Bulletin, 1452, 456. R OCKOW , K. M., H ASKIN , L. A., J OLLIFF , B. L. & F OUNTAIN , D. M. 1997. Constraints on element mobility associated with the conversion of granulite to eclogite along fractures in an anorthositic complex on Holsnøy, Norway. Journal of Metamorphic Geology, 15, 401–418. R UBIE , D. C. 1990. Role of kinetics in the formation and preservation of eclogites. In: C ARSWELL , D. A. (ed.) Eclogite Facies Rocks. Blackie, Glasgow, 111–140. R UBIN , J. N., H ENRY , D. C. & P RICE , J. G. 1993. The mobility of zirconium and other ‘immobile’ elements
during hydrothermal alteration. Chemical Geology, 110, 29–47. S ANDIFORD , M. 1985. The origin of retrograde shear zones in the Napier Complex: implications for the tectonic evolution of Enderby Land, Antarctica. Journal of Structural Geology, 7, 477– 488. S ANDIFORD , M. & P OWELL , R. 1986. Pyroxene exsolution in granulites from Fyfe Hills, Enderby Land, Antarctica: Evidence for 1000 8C metamorphic temperatures in Archaean continental crust. American Mineralogist, 71, 946–954. S ANDIFORD , M. & W ILSON , C. J. L. 1984. The structural evolution of the Fyfe Hills–Khmara Bay region, Enderby Land, East Antarctica. Australian Journal of Earth Sciences, 31, 403– 426. S HERATON , J. W., O FFE , L. A., T INGEY , R. J. & E LLIS , D. J. 1980. Enderby Land, Antarctica—an unusual Precambrian high-grade metamorphic terrain. Australian Journal of Earth Sciences, 27, 1– 18. S HERATON , J. W., T INGEY , R. J., B LACK , L. P., O FFE , L. A. & E LLIS , D. J. 1987. Geology of Enderby Land and western Kemp Land, Antarctica. Bureau of Mineral Resources, Australia, Bulletin, 223. S HIRAISHI , K., E LLIS , D. J., F ANNING , C. M., H IROI , Y., K AGAMA , H. & M OTOYOSHI , Y. 1997. Reexamination of the metamorphic and protolith ages of the Rayner Complex, Antarctic: Evidence for the Cambrian (Pan-African) regional metamorphic event. In: R ICCI , C. A. (ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antartica, Siena, 79–88. S TACEY , J. S. & K RAMERS , J. D. 1975. Approximation of terrestrial lead isotope evolution by a two-stage model. Earth and Planetary Science Letters, 26, 207–221. S VERJENSKY , D. A., H EMLEY , J. J. & D’A NGELO , W. M. 1991. Thermodynamic assessment of hydrothermal alkali feldspar–mica–aluminosilicate equilibria. Geochimica et Cosmochimica Acta, 55, 981– 1104. T AYLOR , S. R. & M C L ENNAN , S. M. 1985. The Continental Crust: its Composition and Evolution: an Examination of the Geochemical Record Preserved in Sedimentary Rocks. Blackwell, Oxford. T HOMPSON , A. B. & T RACY , R. J. 1979. Model systems for anatexis of pelitic rocks II. Facies series melting and reactions in the system CaO–KAlO2 – NaAlO2 – Al2O3 – SiO2 – H2O. Contributions to Mineralogy and Petrology, 70, 429– 438. T SUNOGAE , T., O SANAI , Y., T OYOSHIMA , T., O WADA , M., H OKADA , T. & C ROWE , W. A. 1999. Fluorinerich calcic amphiboles in ultrahigh- temperature mafic granulite from Tonagh Island in the Napier Complex, East Antarctica: Preliminary report. Polar Geoscience, 13, 103– 113. W HITE , S., B URROWS , S. E., C ARRERAS , J., S HAW , N. D. & H UMPHREYS , F. J. 1980. On mylonites in ductile shear zones. Journal of Structural Geology, 2, 175–187. Z HU , C. & S VERJENSKY , D. A. 1992. F– Cl– OH partitioning between biotite and apatite. Geochimica et Cosmochimica Acta, 56, 3435– 3467.
Carbonic fluids in ultrahigh-temperature metamorphism: evidence from Raman spectroscopic study of fluid inclusions in granulites from the Napier Complex, East Antarctica T. TSUNOGAE1,2, M. SANTOSH3, J. DUBESSY4, Y. OSANAI5, M. OWADA6, T. HOKADA7 & T. TOYOSHIMA8 1
Graduate School of Life and Environmental Sciences (Earth Evolution Sciences), University of Tsukuba, Ibaraki 305-8572, Japan (e-mail:
[email protected]) 2
Department of Geology, University of Johannesburg, P.O. Box 524, Auckland Park 2006, Republic of South Africa 3
Faculty of Science, Kochi University, Akebono-cho 2-5-1, Kochi 780-8520, Japan 4
CREGU-G2R UMR 7566, Universite´ Henri Poincare´, Nancy 1, BP 239, 54501 Vandoeuvre-le`s-Nancy Cedex, France
5
Division of Evolution of Earth Environment, Graduate School of Social and Cultural Studies, Kyushu University, Fukuoka 810-8560, Japan
6
Department of Geosphere Sciences, Yamaguchi University, Yamaguchi 753-8512, Japan 7
National Institute of Polar Research, Itabashi, Tokyo 173-8515, Japan
8
Graduate School of Science and Technology, Niigata University, Niigata 950-2181, Japan Abstract: We report the first quantitative compositional data on fluid inclusions in ultrahightemperature (UHT) granulites from the Napier Complex of Enderby Land, East Antarctica. Fluid inclusions in various high-grade minerals such as garnet, orthopyroxene and sapphirine from three UHT localities in the Amundsen Bay area were studied in terms of petrography and microthermometry as well as laser Raman spectroscopy. Measured melting temperatures of inclusions from all the three localities indicate that the trapped fluid phase is dominantly carbonic. Raman analyses confirmed a near pure CO2 composition with only minor dilutants such as N2 (,6.0 mol%), CH4 (,0.3 mol%), and H2O (,0.1 mol%). CH4-bearing fluid associated with sapphirine granulites suggests low oxygen fugacity ( f O2) conditions for the rocks, whereas CH4 was not detected from fluid inclusions in magnetite-bearing high-f O2 garnet granulite. The range of CO2 isochores computed from density measurements in fluid inclusions from the granulites pass through the peak P– T conditions of the Napier metamorphism (T ¼ 1050– 1150 8C, P ¼ 9–11kbar) indicating synmetamorphic nature of the fluids. Inclusions in garnet from Bunt Island coexist with carbonate minerals (magnesite) and graphite along with dense CO2-rich fluid, indicating probable derivation from deep-seated primary magmatic sources. The ubiquitous association of carbonic fluids in the UHT mineral assemblages suggests CO2 influx during extreme crustal metamorphism of the Napier Complex. The carbonic fluid probably played an important role in transporting heat from mantle or mantle-derived magmas and in stabilizing the dry mineral assemblages.
The Napier Complex of Enderby Land, East Antarctica, is a typical example of a Neoarchaean regional metamorphic terrane that underwent T . 1000 8C ultrahigh-temperature (UHT) metamorphism (e.g. Ellis 1980; Harley & Hensen 1990; Harley 1998a; Harley & Motoyoshi 2000; Hokada 2001). The Amundsen Bay area, located in the western part of the Napier Complex, is regarded as the highest grade region of the complex, from where equilibrium sapphirine–quartz assemblage
has been widely reported (e.g. Dallwitz 1968; Ellis 1980; Ellis et al. 1980; Grew 1980, 1982; Motoyoshi & Matsueda 1984; Sheraton et al. 1987; Motoyoshi & Hensen 1989; Harley & Hensen 1990; Osanai et al. 2001a, b; Tsunogae et al. 2002). The orthopyroxene sillimanite–garnet assemblage (e.g. Harley 1985, 1998b; Osanai et al. 1999), osumilite (e.g. Ellis 1980; Ellis et al. 1980; Grew 1982), inverted pigeonite in meta-ironstone (e.g. Grew 1982; Sandiford & Powell 1986; Harley 1987), and
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 317 –332. DOI: 10.1144/SP308.15 0305-8719/08/$15.00 # The Geological Society of London 2008.
318
T. TSUNOGAE ET AL.
high-fluorine phlogopite and pargasite in pelitic and basic granulites (e.g. Motoyoshi & Hensen 2001; Tsunogae et al. 2003a) in this area provide additional evidence for regional UHT metamorphism (Harley 1998a). UHT metamorphism in the lower crust is generally characterized by the occurrence of anhydrous mineral assemblages in various rocks. Activity of H2O during peak metamorphism is therefore regarded to have been buffered toward a low level. Therefore, the composition of fluids associated with metamorphism probably played an important role in stabilizing UHT mineral assemblages. Since Touret (1977) first reported that the transition from amphibolite to granulite facies is associated with a change from H2O-rich to CO2rich fluids, numerous studies on CO2-rich fluid inclusions trapped in granulites have been published (e.g. Santosh & Tsunogae, 2003, and references therein). Recently, CO2-rich fluid inclusions have also been reported from some UHT terranes (e.g. Tsunogae et al. 2002; Sarkar et al. 2003; Santosh et al. 2004; Cuney et al. 2007). Such
carbonic inclusions in UHT minerals are typically characterized by very high density (higher than 1 g cm23). Although the timing and mechanism of entrapment of CO2-rich fluids in granulites has been debated (e.g. Lamb et al. 1987, 1991; Santosh et al. 1991; Cesare et al. 2005), the wide occurrence of CO2-rich fluid inclusions of synmetamorphic nature in granulite-facies rocks has been taken to indicate that carbonic fluids play a major role in lowering the H2O activity and stabilizing the anhydrous mineral assemblages that characterize granulites (e.g. Newton et al. 1980; Santosh 1992). In this paper we discuss the occurrence and composition of fluid inclusions trapped in highgrade minerals such as garnet, orthopyroxene and sapphirine in various UHT rocks from Tonagh Island, Bunt Island and Priestley Peak in the Amundsen Bay area (Fig. 1). Fluid inclusions in UHT rocks from Tonagh and Bunt Islands were earlier examined by Tsunogae et al. (2002, 2003b), and they reported high-density (up to 1.07 g cm23) carbonic fluids associated with the peak UHT metamorphism of the areas. However,
Fig. 1. Location map of Tonagh Island, Bunt Island and Priestley Peak in Amundsen Bay area, East Antarctica.
UHT METAMORPHIC FLUIDS
no quantitative measurement of fluid compositions or identification of solid phases associated with the carbonic fluid has been done so far. In this study, we employed laser Raman spectroscopic analysis and present quantitative compositional data for selected fluid inclusions trapped at the peak UHT metamorphism. The results provide the first quantitative database on fluids involved in extreme crustal metamorphism in the Napier Complex and are hence useful for the evaluation of fluid processes of UHT metamorphism within the continental lower crust.
Geological setting and sample description
319
granulite and garnet-bearing felsic gneiss. Magnetite quartzite, garnet–orthopyroxene gneiss, and ultramafic granulite are minor constituents of the layered gneisses. Peak metamorphic conditions of Tonagh Island were determined as T . 1100 8C using ternary-feldspar equilibrium (Hokada 2001) and phase equilibria on sapphirine granulites (Osanai et al. 1999). Temperatures of 1050– 1100 8C were also estimated using Al solubility in orthopyroxene (Tsunogae et al. 2002). A counterclockwise P–T path has been proposed for the area based on geothermobarometry and phase analysis (Tsunogae et al. 2002). Two lithologies have been examined for fluid inclusion study: sapphirine granulite (sample B98021104A) and garnet–orthopyroxene granulite (sample B98022208C). Sample B98021104A is composed of garnet (Alm48 – 52Pyr46 – 51Grs1 – 2 Sps0 – 1), quartz, mesoperthite, sillimanite, sapphirine (XMg ¼ 0.79–0.81), orthopyroxene (XMg ¼ 0.72– 0.73, XAl 0.20), cordierite (XMg ¼ 0.89–0.90), and secondary biotite. The rock shows weak foliation defined by quartzo-feldspathic layers and garnet-rich layers. Sapphirine and quartz coexist in the quartzofeldspathic layers, although the minerals are separated by a corona of garnet þ sillimanite þ cordierite with minor orthopyroxene (Figs 2a and 3a). This texture suggests that the rock underwent T . 1000 8C UHT metamorphism followed by isobaric cooling to form the retrograde corona through the progress of the FMAS reactions
The Amundsen Bay area in the Napier Complex is composed mainly of layered quartzo-feldspathic, pelitic and psammitic gneisses with subordinate orthopyroxene–quartz–feldspar gneiss, mafic and ultramafic granulites, and magnetite quartzite (Sheraton et al. 1987). P–T estimates of the various lithologies from this region based on different methods such as conventional geothermometry (e.g. Ellis 1980), phase analysis (e.g. Grew 1980; Harley & Hensen 1990; Harley et al. 1990; Harley 1998a), Al solubility in orthopyroxene (e.g. Harley & Motoyoshi 2000), inverted pigeonite (e.g. Harley 1987), oxygen isotope geothermometry (e.g. Farquhar et al. 1996), and ternary feldspar geothermometry (e.g. Hokada 2001) suggest that the rocks were subjected to UHT metamorphism. Available geochronological data suggest that the granulite-facies peak metamorphism occurred during the Neoarchaean (c. 2.5 Ga, e.g. Harley & Black 1997; Hokada et al. 2003). The geological features of the locations examined in this study are summarized below, together with descriptions of the analysed samples. The samples were collected during a field geological survey of the Amundsen Bay area undertaken by the 39th Japanese Antarctic Research Expedition (JARE-39) in 1998. Representative textures of the samples are shown in Figures 2 and 3. Mineral abbreviations are after Kretz (1983). Chemical analyses of representative minerals were performed by electron microprobe analyser (JEOL JXA8621) at the University of Tsukuba under conditions of 20 kV accelerating voltage and 10 nA sample current.
Sample B98022208C has a mineral assemblage of garnet (Alm79 – 80Pyr11 – 12Grs7 – 8Sps1 – 2), orthopyroxene (XMg ¼ 0.34 –0.37), quartz, and magnetite (Fig. 2b). The foliation of the gneiss is defined by oriented quartz and orthopyroxene. Garnet and orthopyroxene are present as coarse-grained (up to 1 cm) porphyroblasts, probably as products of prograde metamorphism. Aggregates of fine-grained orthopyroxene with minor quartz are present occasionally around coarse-grained garnet and orthopyroxene, and are probably a product of a later deformation event.
Tonagh Island
Bunt Island
The detailed regional geological framework and structural characteristics of Tonagh Island have been given by Osanai et al. (2001a). The area is dominantly composed of layered gneisses of various bulk chemical composition. Orthopyroxene-bearing quartzo-feldspathic gneiss is the most abundant lithology. It is commonly interlayered with mafic
The island is underlain by layered gneisses containing garnet-bearing and orthopyroxene-bearing quartzo-feldspathic gneisses, mafic and ultramafic granulites, and garnet–sillimanite (pelitic) gneiss (Osanai et al. 2001b). The sapphirine- and osumilite-bearing aluminous granulite selected for fluid inclusion study occurs between boudinaged
Spr þ Qtz ! Grt þ Sil þ Crd Spr þ Qtz ! Opx þ Sil þ Crd:
(1) (2)
320
T. TSUNOGAE ET AL.
Fig. 2. Photomicrographs of granulites discussed in this paper. Mineral abbreviations are after Kretz (1983). (a) Sapphirine þ quartz assemblage in pelitic granulite (sample B98021104A, Tonagh Island). Sillimanite þ garnet corona is formed between sapphirine and quartz as a result of isobaric cooling after the peak UHT metamorphism. (b) Orthopyroxene þ garnet assemblage in sample B98022208C (Tonagh Island). (c) Sapphirine þ garnet þ osumilite assemblage in quartz-absent portion of Mg–Al-rich rock (sample BF1, Bunt Island). (d) Sapphirine including quartz in the matrix of K-feldspar (sample TS98022407, Priestley Peak). Retrograde sillimanite and garnet are also present between this sapphirine–quartz pair.
ultramafic granulite and sillimanite–garnet–sapphirine gneiss, which in turn is surrounded by garnet- and orthopyroxene-bearing quartzo-feldspathic gneisses. Osanai et al. (2001b) estimated the UHT peak metamorphic condition of Bunt Island as T . 1030 8C and P , 9.8 kbar based on a petrogenetic grid and the THERMOCALC program. On the basis of detailed petrographic observations and thermodynamic considerations in the KFMAS system, they proposed that the garnet–K-feldspar–sillimanite assemblage is replaced by sapphirine–garnet– K-feldspar and sapphirine2garnet2osumilite assemblages as a result of a pressure drop at almost constant temperature through the progress of the reactions
Grt þ Sil ! Spr þ Qtz
(3)
Grt þ Spr þ Kfs þ Qtz ! Osm:
(4)
Sample BF1 is composed of orthopyroxene XAl ¼ 0.24–0.25), quartz, (XMg ¼ 0.68–0.69, garnet (Alm43 – 44Pyr53 – 54Grs1 – 2Sps0 – 1), sapphirine (XMg ¼ 0.76–0.80), sillimanite, K-feldspar, and osumilite. The rock comprises orthopyroxene– quartz–osumilite-rich layers and sapphirine– garnet–osumilite-rich aluminous layers (Fig. 2c). Garnet in the sample occurs as medium- to coarsegrained (0.3–4.3 mm) poikiloblasts in the aluminous layers. Common intergrowth of garnet and sapphirine suggests its formation during a high-grade event. Fluid inclusions are commonly trapped in this type of garnet. A fine-grained sapphirine þ quartz assemblage along the grain boundary of orthopyroxene suggests peak UHT metamorphism at T . 1000 8C (Fig. 3b). Orthopyroxene and quartz in felsic layers are coarse-grained (up to 2 mm) and occasionally show granoblastic texture.
UHT METAMORPHIC FLUIDS
321
Priestley Peak Priestley Peak is located about 10 km south of Tonagh Island (Fig. 1). No detailed field geological survey has been carried out so far on this locality. The study area, which corresponds to the coastal region of the exposure, is composed of layered gneisses of quartzo-feldspathic, mafic and pelitic compositions. The examined pelitic gneiss sample TS98022407 is composed of mesoperthite, quartz, sillimanite, garnet (Alm42 – 43Pyr54 – 55Grs1 – 2Sps0 – 1), and spinel sapphirine (XMg ¼ 0.78–0.81), (XMg ¼ 0.56–0.57). Sapphirine is occasionally poikiloblastic and contains inclusions of quartz and mesoperthite (Figs 2d and 3c). The boundary between the sapphirine and quartz is sharply defined (Fig. 3c). This texture confirms that the peak metamorphic temperature of this area is higher than 1000 8C, which is consistent with the conditions for the other two localities.
Fluid inclusions Analytical procedure
Fig. 3. Back-scattered electron images showing detailed texture of sapphirine–quartz association in UHT metamorphic rocks from the Napier Complex discussed in this paper. (a) Garnet þ sillimanite þ orthopyroxene corona between sapphirine and quartz suggesting postpeak isobaric cooling history (sample B98021104A, Tonagh Island). (b) Fine-grained sapphirine þ quartz assemblage along the grain boundary of porphyroblastic orthopyroxene (sample BF1, Bunt Island). (c) Sapphirine þ quartz assemblage in quartzo-feldspathic layer of pelitic granulite (sample TS98022407, Priestley Peak).
Fluid inclusions were studied in doubly polished thin wafers (c. 150–200 mm in thickness) prepared from representative rock samples. The nature of occurrence of inclusions, their distribution pattern, shape, size and phase categories were carefully studied and documented under a petrological microscope at varying magnifications following the techniques outlined by Touret (2001) and Van den Kerkhof & Hein (2001). Photomicrographs of representative fluid inclusions are shown in Figure 4. Microthermometric measurements were performed with US Geological Survey and Linkam heating– freezing systems at the University of Tsukuba. Calibration was undertaken with synthetic standard materials supplied by Fluid Inc., Denver. Heating rates of the samples are 18C min21 for melting and 5 8C min21 for homogenization temperatures. Repeated microthermometric measurements indicate that the precision of microthermometric results reported in this study is within +0.2 8C. The results of microthermometric measurements are shown in Figure 5 and summarized in Table 1. In all the examined cases, the CO2 homogenizes into the liquid phase. Fluid densities and isochores were calculated using the computer program ‘FLINCOR’ developed by Brown (1989) based on the equation and thermodynamic data of Brown & Lamb (1986). Laser Raman spectroscopy provides a powerful tool for the identification and quantification of fluids trapped in minerals (e.g. Wopenka & Pasteris 1986; Pasteris et al. 1988; Dubessy et al. 1989). The molar
322
T. TSUNOGAE ET AL.
Fig. 4. Photomicrographs of representative fluid and solid inclusions reported in this paper. (a) Primary cluster of carbonic fluid inclusions (FI) in quartz associated with sapphirine in sample B98021104A. (b) Pseudosecondary carbonic inclusions within garnet in sample B98022208C. (c) Aggregates of about 10–20 primary carbonic inclusions in garnet in sample BF1. (d) Magnesite (Mgs) and graphite (Gr) associated with a high-density pure CO2 inclusion in garnet in sample BF1. Fine-grained rutile (Rt) is also present with the inclusion. (e) Pseudosecondary fluid inclusion trail in garnet in sample TS98022407. (f) Isolated primary carbonic inclusions in sillimanite enclosed in garnet in sample TS98022407.
ratio of gas phases in single fluid inclusions was analysed using a Dilorw Labram type Raman spectrometer at CREGU, Nancy, with a Notchw filter and a CCD detector cooled at 230 8C. The exciting radiation at 514.532 nm is provided by an Arþ laser
(type 2020, Spectraphysicsw). Laser power is around 60 mW at the focused point. Spatial resolution is around a few microns, although it depends on the depth of the inclusions. Raman spectroscopic analysis was carried out on relatively
UHT METAMORPHIC FLUIDS
323
Fig. 5. Histograms showing the distribution of melting (a) and homogenization (b) temperatures of fluid inclusions in the examined UHT granulites. Most microthermometric data for samples from Tonagh and Bunt Islands are taken from Tsunogae et al. (2002, 2003b). (See Table 1 for abbreviations.)
large inclusions (more than 10 mm) to obtain high Raman signals for quantitative analyses. Raman data were obtained at room temperature for singlephase inclusions and at 150 8C for two-phase (H2O þ CO2) inclusions after homogenizing the inclusion fluids using a Linkam heating–cooling stage fixed on the microscope of the Raman microprobe. The analysis was carried out for granulite samples from two localities, Tonagh and Bunt Islands. Raman spectra of CO2, N2 and CH4 for representative fluid inclusions are shown in Figure 6a and b. The H2O spectrum was not obtained in this study because of its very low signals. No other gases, such as H2, CO, H2S, SO2 or C2H6, were detected. The detailed procedure of quantitative analysis of fluid inclusions has been discussed by Dubessy et al. (1989). Briefly, estimates of relative molar ratios of fluid species in fluid inclusions have been made based on peak ratios using equation (2) of Dubessy et al. (1989). The V– X properties of fluids in the C–O –H – N– S system were deduced mainly from roomtemperature measurements. Precision of the Raman spectroscopic analysis of fluid inclusions has been discussed in detail by Wopenka & Pasteris (1986). The results of quantitative Raman analyses are summarized in Table 2.
Results Tonagh Island. Fluid inclusions in sapphirine granulite (sample B98021104A) are present in quartz and rarely in sapphirine. Those in quartz occur as isolated clusters and have irregular shapes that
correspond to primary inclusions (Fig. 4a), whereas those in sapphirine occur together with an unknown solid mineral. Although the inclusions show consistent melting temperatures of 256.8 to 256.4 8C in both minerals, those in quartz show a slightly lower range of homogenization temperature (231.5 to 213.2 8C) as compared with the inclusions in sapphirine (214.3 8C) (Tsunogae et al. 2002). Raman analysis was performed on relatively large primary inclusions in quartz. The results show that the trapped fluid is nearly pure CO2 (99.9 mol%) (Fig. 6a) with minor impurities such as CH4 (less than 0.1 mol%) and H2O (0.1 mol%). Fluid inclusions in sapphirine could not be analysed because the grain size is too small. For the inclusions in this sample, CO2 densities of 1.00– 1.06 g cm23 were computed. Coarse-grained garnet, orthopyroxene and matrix quartz in sample B98022208C contain numerous fluid inclusions. Fluid inclusions are particularly abundant in garnet as pseudosecondary clusters of 20 –100 inclusions (Fig. 4b). In contrast, most of the fluid inclusions of pseudosecondary origin in orthopyroxene are now empty cavities, filled with secondary minerals. These mineral inclusions are 2–10 mm in size and have an elongated tubular shape. Fluid inclusions in quartz are relatively smaller (generally less than 4 mm), have ovoidal shape, and are generally distributed along within-grain trails as a pseudosecondary type. All the inclusions show melting temperatures of 257.2 to 256.4 8C, and have been confirmed as CO2-rich by Raman analysis (98.0– 99.8 mol% CO2). Other fluid phases such as N2 and H2O
324
Table 1. Microthermometric data for representative fluid inclusions in granulites from the Napier Complex Locality and sample number Tonagh Island B98021104A
Bunt Island BF1
Priestley Peak TS98022407
Type of inclusions
Morphology
Melting temperature
Homogenization temperature
Min.
Max.
Average*
Min.
Max.
Average*
Density (g cm23)
Qtz Spr Qtz Grt Opx
PR, PS PR PS PS PS
IR, NC IR NC NC NC
256.8 256.7 257.2 256.7 256.6
256.4 256.4 256.6 256.4 256.6
256.6 + 0.1 256.6 + 0.1 256.9 + 0.1 256.6 + 0.1 256.6
231.5 214.3 229.1 224.8 23.7
213.2 214.3 2.5 4.2 23.7
223.6 + 3.9 214.3 226.3 + 1.7 212.9 + 7.4 23.7
1.03– 1.06 1.00 1.05– 1.07 0.96– 1.03 0.95
Grt (high-density) Grt (low-density) Qtz (high-density) Qtz (low-density) Opx Spr
PR PR PR PR PS PS
RS, IR RS, IR RS RS IR NC
257.7 257.8 257.3 257.4 257.6 257.7
257.3 256.8 257.0 257.2 257.6 257.4
257.5 + 0.1 257.5 + 0.2 257.2 + 0.1 257.3 + 0.1 257.6 257.6 + 0.1
235.4 212.0 24.8 6.3 223.6 22.7
224.6 10.8 1.1 24.7 11.5 10.2
229.5 + 2.3 25.1 + 4.5 21.7 + 2 14.6 + 4.1 21.0 + 13.4 5.4 + 3.2
1.06– 1.08 0.93– 0.98 0.93– 0.95 0.79– 0.86 0.84– 1.01 0.87– 0.91
Grt Grt
PR S
RS RS
256.8 257.0
256.7 256.5
256.8 + 0.1 256.7 + 0.2
235.5 211.3
233.9 þ14.6
234.6 + 0.7 þ4.5 + 6.0
1.09– 1.10 0.86– 0.94
PR, primary; PS, pseudosecondary; S, secondary; RS, rounded shape; IR, irregular shape; NC, negative crystal. *Error indicates standard deviation.
T. TSUNOGAE ET AL.
B98022208C
Host mineral
UHT METAMORPHIC FLUIDS
325
Fig. 6. Raman spectra of representative fluid and solid inclusions discussed in this study. (a) Primary CO2-rich inclusion within quartz in sample B98021104A (Fig. 4a). (b) N2 and CH4 dissolved in CO2-rich fluid within garnet in sample BF1. (c) Magnesite (Fig. 4d) trapped together with a high-density CO2 inclusion within garnet in sample BF1. (d) Graphite (Fig. 4d) associated with a high-density CO2 inclusion within garnet in sample BF1.
were also detected from fluid inclusions in garnet and quartz, although they are minor (less than 2.0 mol% and 0.1 mol%, respectively). It should be noted that CH4 was not detected from fluid inclusions in this sample. This is probably related to the higher oxidation state as indicated by the occurrence of magnetite in this sample. For the inclusions in the sample, CO2 densities of 0.96–1.03 g cm23 are computed for pseudosecondary inclusions in garnet, 0.95 g cm23 for pseudosecondary inclusions in orthopyroxene, and 1.05–1.07 g cm23 for pseudosecondary inclusions in quartz (Tsunogae et al. 2002). Bunt Island. Fluid inclusions are present in quartz, garnet, sapphirine and orthopyroxene in sample BF1. Those in other minerals such as osumilite and K-feldspar are rare and too small to analyse. Fluid inclusions in quartz and garnet occur as isolated inclusions or local clusters representing a primary type (Fig. 4c). Fluid inclusions in sapphirine and orthopyroxene occurs as an intra-crystal planar array, which corresponds to pseudosecondary type inclusions. Melting temperatures of all the fluid inclusions vary from 257.8 to 256.8 8C, close to the triple point of pure CO2 (Tsunogae et al. 2003b). Laser Raman analysis of fluid inclusions in sapphirine and garnet indicates that the slight
depression in the melting temperatures below 256.6 8C is probably due to the presence of traces of additional fluid components such as N2 (up to 6.0 mol%) and CH4 (up to 0.3 mol%) (Fig. 6b). Homogenization temperatures of fluid inclusions in garnet define two peaks, around 229.5 + 2.3 8C (1.06–1.08 g cm23) and 25.1 + 4.5 8C (0.93–0.98 g cm23), and such marked difference in fluid densities within the same group of synchronous inclusions in garnet probably reflects density reversal of originally high-density inclusions. We therefore infer that the inclusions showing lower homogenization temperature may record the history of the early (probably peak) stage of metamorphism. The homogenization temperatures of fluid inclusions in quartz indicate a peak at around 21.7 + 2.0 8C (0.93–0.95 g cm23), which is nearly consistent with the low-density fluid inclusions in garnet. On the other hand, some inclusions show very high homogenization temperatures compared with the others, showing a broad peak at around þ14.6 + 4.1 8C (0.79–0.86 g cm23). These inclusions were obviously trapped at a later stage of the exhumation history of the rock and do not preserve the peak metamorphic history. The homogenization temperatures of pseudosecondary fluid inclusions in sapphirine and orthopyroxene show a wide variation, from
326
T. TSUNOGAE ET AL.
Table 2. Molar abundance of fluids trapped in fluid inclusions in UHT rocks from the Napier Complex Locality and sample number Tonagh Island B98021104A B98022208C
Bunt Island BF1
Inclusion number
Host mineral
Type of inclusions
Tm (8C)
Th (8C)
Q1-2(q769) Q1-2(q760) Q1(q712) Q2(q713) G1(q746)
Qtz Qtz Qtz Qtz Grt
PR PR PR PR PR
256.6 256.9 257.0 256.8 256.7
223.5 225.5 228.4 226.5 216.5
99.9 99.9 98.2 99.7 98.0
,0.1 n.d. n.d. n.d. n.d.
n.d. n.d. 1.8 0.3 2.0
0.1 0.1 – – –
132(q754) 7-2(q758) 7(q719) 16(q720)
Spr Grt Grt Grt
PR PR PR PR
257.7 þ4.8 257.4 23.5 – – 257.4 230.6
94.0 99.8 96.0 99.3
n.d. 0.1 0.3 n.d.
6.0 n.d. 3.7 0.7
n.d. ,0.1 – –
CO2 CH4 N2 H2O (mol%) (mol%) (mol%) (mol%)
PR, primary; Tm, melting temperature; Th, homogenization temperature; –, not analysed; n.d., not detected.
223.6 to þ11.5 8C (0.84–1.01 g cm23), although only a limited number of inclusions were analysed. In rare cases, the carbonic fluid inclusion in garnet is closely associated with solid phases (Fig. 4d). The solid inclusion occurs as euhedral to irregular crystals of c. 5– 20 mm in size, and has been identified as magnesite by Raman spectroscopic analysis (Fig. 6c). Such carbonatebearing inclusions are also occasionally associated with graphite (Figs 4d and 6d). Although the graphite is very fine grained (less than 5 mm), smooth crystal surface can be observed at high magnification. The Raman spectrum shows a well-defined high-intensity peak at around 1560 cm21 as well as a low-intensity broad peak at around 1340 cm21 (Fig. 6d), suggesting a lower degree of crystallinity. To the best of our knowledge, this is the first report of carbonate and graphite in association with CO2 within fluid inclusions from UHT rocks and has important implications for the role and source of CO2 in extreme crustal metamorphism. Priestley Peak. No fluid inclusion data have been published so far from this locality. Fluid inclusions occur dominantly in garnet in sample TS98022407. They are texturally secondary, forming trails across grain boundaries of the minerals (Fig. 4e). Those trapped together with sillimanite in the core of garnet occur as isolated inclusions as texturally primary phase (Fig. 4f). Melting temperatures of all the fluid inclusions are close to 256.6 8C (257.0 to 256.5 8C). Although Raman analysis has not been performed for the fluid inclusions, the melting temperature range suggests that the captured fluid is near pure CO2. Homogenization temperatures of the secondary inclusions fall within the range of 211.3 to þ14.6 8C (0.86– 0.94 g cm23). In contrast, the homogenization temperatures of
primary inclusions trapped together with sillimanite indicate a range from 235.5 to 233.9 8C, which corresponds to very high-density carbonic fluid of 1.09 –1.10 g cm23.
Discussion Petrography and microthermometric measurements of fluid inclusions in high-grade minerals such as garnet, orthopyroxene and sapphirine from three localities in the Napier Complex indicate that fluids trapped in the minerals formed during the UHT metamorphism are dominantly CO2. Laser Raman spectroscopic analysis of these fluid inclusions from Tonagh and Bunt Islands confirmed the CO2-rich composition with minor contents of dilutants such as CH4, N2 and H2O; N2 varies from 0 to 6.0 mol%, CH4 from 0 to 0.3 mol%, and H2O from 0 to 0.1 mol%. The occurrence of near pure CO2 in all the samples examined in this study from different localities in the Amundsen Bay area suggests that the ambient fluid during the major thermal events in the highest-grade region of the Napier Complex was carbonic in nature. The result is consistent with the common occurrences of anhydrous mineral assemblages throughout the study area. This is the first quantitative investigation of fluid composition in high-grade minerals associated with UHT metamorphism of the complex, and our results have important implications in evaluating the nature and composition of fluids associated with Neoarchaean extreme thermal metamorphism in the lower crust. Isochores of CO2-rich fluid inclusions calculated from density measurements in fluid inclusions in granulites from three locations are shown in Figure 7 together with the broad P –T windows and trajectories obtained from geothermobarometry
UHT METAMORPHIC FLUIDS
and mineral phase equilibria. Although we did not estimate P –T conditions from the analysed rocks in this study, occurrences of sapphirine þ quartz assemblage in samples B98021104A, BF1 and TS98022407 from the three localities suggest that the samples underwent peak UHT metamorphism as presented in Figure 7. A counterclockwise P –T path is proposed for Tonagh Island because of the common occurrence of textures suggesting reaction (1) given above (Tsunogae et al. 2002). In contrast, near-isothermal decompression following a clockwise trajectory has been inferred for Bunt Island (Osanai et al. 2001b). Figure 7 demonstrates that the isochores of carbonic inclusions are almost consistent with the peak P –T conditions of the Napier Complex (1050– 1150 8C, 9–11 kbar). The texturally primary nature of most of the fluid inclusions investigated in this study suggests that the fluid capture might have occurred at the high-grade stage and that the carbonic fluids represent traces of the ambient fluid present during extreme crustal metamorphism of these rocks. Such ‘ultrahighdensity’ carbonic fluids have also been reported from Palaeoproterozoic granulites from In Ouzzal, Algeria (Cuney et al. 2007), the Grenvillian Eastern Ghats belt, India (Sarkar et al. 2003), Pan-African granulite complexes in India and Sri Lanka (e.g. Fonarev et al. 2001; Ohyama et al. 2008; Tsunogae et al. 2008) and Neoarchaean granulites in southern India (Santosh & Tsunogae 2003). The occurrence of high-density carbonic fluid trapped as an early phase in high-grade minerals is therefore regarded as a common feature of lower crustal rocks that underwent extreme thermal metamorphism during various stages in Earth’s history. Slightly low-density values of some carbonic fluids (e.g. sample BF1), although texturally these inclusions are primary or pseudosecondary, probably resulted from density reversal during the retrograde stage. Such density reversal of carbonic fluid during rapid uplift along a nearisothermal decompression path has been reported from other UHT terranes also (e.g. Tsunogae & van Reenen 2007). It is interesting to note that, although some inclusions in the Bunt Island sample underwent a significant effect of density reversal, the fluid density of the Tonagh Island samples has not been reset. This is probably due to the difference in the styles of the P–T trajectory between the two localities. Significant volume change of Bunt rocks as a result of near-isothermal decompression might have formed microcracks in minerals and given rise to retrograde fluid leakage. On the other hand, Tonagh Island rocks generally followed a counterclockwise P–T path almost parallel to calculated isochores, and therefore underwent a less significant effect of later volume change.
327
The role of CO2 as the ambient fluid species during UHT metamorphism is a topic that needs to be evaluated through detailed study in future. Harley (2004) observed, based on CO2 content in natural cordierite (Harley et al. 2002), that the calculated channel XCO2 values of cordierite lie below the fluid saturation value for the likely P – T conditions and the calculated CO2 activity values, and hence concluded that these values are less than those required for fluid saturation during cordierite formation at or near peak UHT conditions in the Napier Complex. Therefore, Harley (2004) considered that a free carbonic fluid may not have been present during UHT metamorphism of the complex. These results conflict with our observation that carbonic fluids are trapped as a texturally primary phase in UHT minerals. This discrepancy is probably because the cordierite discussed by Harley (2004) was derived from prograde dehydration melting and all free CO2 in that case might have been incorporated into the cordierite channel. However, on the basis of detailed field and petrological observation, any dominant melting texture accompanying cordierite formation has not been observed from the study area. Cordierite occurs only as a later product around fine-grained retrograde corona garnet in sample B98021104A. Therefore, cordierite was not a probable stable mineral at the peak UHT stage of the study area. This observation is also true with regard to the formation of cordierite mostly as a late-stage mineral, after peak UHT conditions in many UHT granulite terranes, such as the Palaeoproterozoic UHT granulites of North China (Santosh et al. 2007) and the late Neoproterozoic UHT granulites in southern India (Tateishi et al. 2004; Ishii et al. 2006; Shimpo et al. 2006; Tadokoro et al. 2007). Therefore, we consider that a free carbonic fluid was present at the peak UHT stage in the present case. Our Raman analysis indicates that nitrogen and methane occur as minor dilutant phases in some of the UHT rocks in the Napier Complex. This is consistent with the slight depression in melting temperature of CO2-rich inclusions from 256.6 8C as discussed in previous studies (see Roedder 1984). Minor CH4 dissolved in carbonic fluids in sapphirine granulites from Tonagh Island (sample B98021104A) and Bunt Island (sample BF1) suggests that oxygen fugacity (f O2) conditions during high-grade metamorphism were probably low. The sapphirine þ quartz assemblage in the rocks is therefore regarded as an indicator of T . 1000 8C UHT metamorphism at a low fO2 state (e.g. Kelsey et al. 2004). In contrast, carbonic fluids in magnetite-bearing garnet granulite
328
T. TSUNOGAE ET AL.
Fig. 7. P– T diagrams showing representative CO2 isochores of primary fluid inclusions in: 1, sillimanite from Priestley Peak (sample TS98022407); 2, garnet from Bunt Island (sample BF1); 3, quartz from Tonagh Island (sample B98021104A); 4, sapphirine from Tonagh Island (sample B98021104A); 5, garnet from Bunt Island (sample BF1); 6, quartz from Bunt Island (sample BF1). P –T paths (arrows) for Tonagh Island (T) and Bunt Island (B) are taken from Tsunogae et al. (2002) and Osanai et al. (2001b), respectively. FMAS petrogenetic grid is modified after Harley (1998a). Open box indicates approximate P– T condition of the study area. Numbers in parentheses indicate reactions discussed in the text.
from Tonagh Island (sample B98022208C) show no trace of CH4. Both CH4 and N2 might have formed as byproducts of the reaction of H2O with NH4 released from the breakdown of biotite at the prograde stage (Dubessy & Ramboz, 1986). Mixing of these fluids with pure CO2 can yield the fluid compositions obtained in the present study. The laser Raman data presented in this study further indicate the coexistence of carbonate minerals (magnesite) and graphite along with CO2-rich fluid inclusions in sample BF1. The intimate association of near pure CO2-rich fluids with carbonate minerals is an important feature observed
for the first time from the Napier Complex. Bolder-Schrijver et al. (2000) have described similar CO2 –carbonate relationships from a sapphirine granulite of the Highland Complex of Sri Lanka. They regarded the carbonate as daughter crystals within CO2 inclusions because the inclusions showed a relatively constant fluid : solid ratio. However, our texture is different from the Sri Lankan fluid –solid inclusions because the examined CO2 – calcite inclusions show a highly variable fluid : solid volume ratio of c. 5 : 1 to 1 : 0. We therefore consider that magnesite in Figure 4d is not a daughter crystal. Even if we consider fluid
UHT METAMORPHIC FLUIDS
loss by partial leakage, the widely varying size range of the carbonates occurring within the same generation of inclusions precludes them from being daughter crystals. Such carbonates in association with very high-density CO2 inclusions with various fluid : solid volume ratio were also noticed within garnet from granulites of the Palghat– Cauvery Shear Zone system in southern India by Tsunogae et al. (2008) within garnet. Graphite associated with CO2-rich fluid inclusions has been reported in previous studies (e.g. Satish-Kumar 2005, and references therein). The Raman spectrum of the graphite inclusion in sample BF1 shows a well-defined high-intensity peak at around 1560 cm21 as well as a low-intensity broad peak at around 1340 cm21 (Fig. 6d), suggesting a lower degree of crystallinity. Previous studies of graphite occurring within carbonic fluid inclusions indicated similar Raman signals, suggesting that the graphite might have been formed at temperatures below the peak metamorphic conditions by precipitation from carbonic fluid (Pasteris & Chou 1998; Satish-Kumar 2005). Relatively high homogenization temperature (i.e. low density) of the graphite-bearing inclusions (23.5 8C) as compared with other solid-free inclusions (235.4 to 224.6 8C) in the same sample supports the graphite precipitation model. Although several possible scenarios to crystallize graphite from carbonic fluid in the C– O– H system during retrograde metamorphism have been proposed (e.g. simple cooling, hydrogen diffusion, hydration reaction, and oxyexsolution as summarized by Satish-Kumar 2005), none of the models can explain the variation in modal abundance of graphite in sample BF1 (c. 0– 5%). Further carbon isotopic studies are required to address the problem of the origin of the graphite within the fluid inclusions. The close association of CO2 and carbonates in these rocks, and the lack of carbonate lithologies adjacent to the examined rocks suggest that the two phases infiltrated together from an external source during UHT metamorphism. The most probable explanation for the CO2 – carbonate + graphite association within the fluid inclusions examined in this study is that the carbonic fluids were probably derived from deep-seated magmatic sources where the CO2 and carbonates were transported by carbonate magmas (see Touret 1985). The CO2 was released upon freezing of the magmas after emplacement, and the fluids along with some of the suspended carbonate minerals infiltrated and became entrapped within the metamorphic minerals that equilibrated within the UHT granulites (Tsunogae et al. 2007). Graphite might have been precipitated from the carbonic fluid at a later stage and formed along with relatively low-density
329
inclusions. The fluid, possibly derived from asthenospheric upwelling followed by subsequent mantle degassing and partial melting, might have played an important role in transporting heat from mantle and stabilizing dry mineral assemblages at a low H2O activity state. We suggest that the CO2 –carbonate + graphite association in fluid inclusions within the UHT granulites is a key indicator for the role of mantle-derived fluid in providing the heat and fluid input for extreme crustal metamorphism. We express our sincere thanks to the members of JARE-39 and the crew of the icebreaker Shirase for giving us the opportunity for geological field investigation of the Napier Complex, and for their helpful support. Especially we thank K. Shibuya, K. Moriwaki, K. Shiraishi and Y. Motoyoshi at the NIPR for their arrangement of our fieldwork and encouragement. The analytical part of this study carried out when the first author visited Nancy in 2004 and 2005. T. Tsunogae thanks Universite´ Henri Poincare´ and the Geology Department of the University of Johannesburg for facilities and support. Special thanks are due to J. L. R. Touret for his valuable comments on our fluid inclusion data. N. Nishida is acknowledged for his assistance with microprobe analyses. This is a contribution to the Grant-in-Aid from the Japanese Ministry of Education, Sports, Culture, Science and Technology to T. Tsunogae (No. 17340158, 20340148) and M. Santosh (No. 17403013). We acknowledge constructive reviews by H. Kagi, M. Satish Kumar and J.-M. Huizenga.
References B OLDER -S CHRIJVER , L. J. A., K RIEGSMAN , L. M. & T OURET , J. L. R. 2000. Primary carbonate/CO2 inclusions in sapphirine-bearing granulites from Central Sri Lanka. Journal of Metamorphic Geology, 18, 259– 269. B ROWN , P. E. 1989. FLINCOR: A microcomputer program for the reduction and investigation of fluid-inclusion data. American Mineralogist, 74, 1390– 1393. B ROWN , P. E. & L AMB , W. M. 1986. Mixing of H2O – CO2 in fluid inclusions: geobarometry and Archean gold deposits. Geochimica et Cosmochimica Acta, 50, 847– 852. C ESARE , B., M ELI , S., N ODARI , L. & R USSO , U. 2005. Fe3þ reduction during biotite melting in graphitic metapelites: another origin of CO2 in granulites. Contributions to Mineralogy and Petrology, 149, 129– 140. C UNEY , M., C OULIBALY , Y. & B OIRON , M. C. 2008. High-density early CO2 fluids in the ultrahightemperature granulites of Ihouhaouene (In Ouzzal, Algeria). Lithos, 96, 402– 414. D ALLWITZ , W. B. 1968. Coexisting sapphirine and quartz in granulite from Enderby Land, Antarctica. Nature, 219, 476–477.
330
T. TSUNOGAE ET AL.
D UBESSY , J. & R AMBOZ , C. 1986. The history of organic nitrogen from early diagenesis to amphibolite facies: mineralogical, chemical, mechanical, and isotopic implications. In: 5th International Symposium on Water–Rock Interaction, Reykjavik, Extended Abstracts, 170–174. D UBESSY , J., P OTY , B. & R AMBOZ , C. 1989. Advances in C–O–H–N–S fluid geochemistry based on microRaman spectroscopic analysis of fluid inclusions. European Journal of Mineralogy, 1, 517–534. E LLIS , D. J. 1980. Osumilite–sapphirine–quartz granulites from Enderby Land, Antarctica: P –T conditions of metamorphism, implications for garnet– cordierite equilibria and the evolution of the deep crust. Contributions to Mineralogy and Petrology, 74, 201 –210. E LLIS , D. J., S HERATON , J. W., E NGLAND , R. N. & D ALLWITZ , W. B. 1980. Osumilite–sapphirine– quartz granulites from Enderby Land, Antarctica: Mineral assemblages and reactions. Contributions to Mineralogy and Petrology, 72, 123– 143. F ARQUHAR , J., C HACKO , T. & E LLIS , D. J. 1996. Preservation of oxygen isotope compositions in granulites from Northwestern Canada and Enderby Land, Antarctica: implications for high-temperature isotopic thermometry. Contributions to Mineralogy and Petrology, 125, 213– 224. F ONAREV , V. I., S ANTOSH , M., F ILMONOV , M. B. & V ASIUKOVA , O. V. 2001. Pressure–temperature– fluid history and exhumation path of a Gondwana fragment: the Trivandrum Granulite Block, southern India. Gondwana Research, 4, 615– 616. G REW , E. S. 1980. Sapphirine þ quartz association from Archean rocks in Enderby Land, Antarctica. American Mineralogist, 65, 821– 836. G REW , E. S. 1982. Osumilite in the sapphirine–quartz terrane of Enderby Land, Antarctica: implications for osumilite petrogenesis in the granulite facies. American Mineralogist, 67, 762–787. H ARLEY , S. L. 1985. Garnet– orthopyroxene bearing granulites from Enderby Land, Antarctica: metamorphic pressure– temperature– time evolution of the Archean Napier Complex. Journal of Petrology, 26, 819– 856. H ARLEY , S. L. 1987. A pyroxene-bearing meta-ironstone, other pyroxene-granulites from Tonagh Island, Enderby Land, Antarctica: further evidence for very high temperature (.980 8C) Archaean regional metamorphism in the Napier Complex. Journal of Metamorphic Geology, 5, 341–356. H ARLEY , S. L. 1998a. On the occurrence and characterization of ultrahigh-temperature crustal metamorphism. In: T RELOAR , P. J. & O’B RIEN , P. J. (eds) What Drives Metamorphism and Metamorphic Reactions? Geological Society, London, Special Publications, 138, 81–107. H ARLEY , S. L. 1998b. Ultrahigh temperature granulite metamorphism (1050 8C, 12 kbar) and decompression in garnet (Mg70)– orthopyroxene–sillimanite gneisses from the Rauer Group, East Antarctica. Journal of Metamorphic Geology, 16, 541– 562. H ARLEY , S. L. 2004. Extending our understanding of ultrahigh-temperature crustal metamorphism. Journal of Mineralogical and Petrological Sciences, 99, 140– 158.
H ARLEY , S. L. & B LACK , L. P. 1997. A revised Archaean chronology for the Napier Complex, Enderby Land, from SHRIMP ion-microprobe studies. Antarctic Science, 9, 74– 91. H ARLEY , S. L. & H ENSEN , B. J. 1990. Archaean and Proterozoic high-grade terranes of East Antarctica (40– 808E): a case study of diversity in granulite facies metamorphism. In: A SHWORTH , J. R. & B ROWN , M. (eds) High-temperature Metamorphism and Crustal Anatexis. Kluwer, Dordrecht, 320–370. H ARLEY , S. L. & M OTOYOSHI , Y. 2000. Al zoning in orthopyroxene in a sapphirine quartzite: evidence for .1120 8C UHT metamorphism in the Napier Complex, Antarctica, and implications for the entropy of sapphirine. Contributions to Mineralogy and Petrology, 138, 293–307. H ARLEY , S. L., H ENSEN , B. J. & S HERATON , J. W. 1990. Two-stage decompression in orthopyroxene– sillimanite granulites from Forefinger Point, Enderby Land, Antarctica: implications for the evolution of the Archaean Napier Complex. Journal of Metamorphic Geology, 8, 591– 613. H ARLEY , S. L., T HOMPSON , P., B UICK , I. S. & H ENSEN , B. L. 2002. Cordierite as a sensor of fluid conditions in high-grade metamorphism and crustal anatexis. Journal of Metamorphic Geology, 20, 71–86. H OKADA , T. 2001. Feldspar thermometry in ultrahightemperature metamorphic rocks: Evidence of crustal metamorphism attaining 1100 8C in the Archean Napier Complex, East Antarctica. American Mineralogist, 86, 932 –938. H OKADA , T., M ISAWA , K., S HIRAISHI , K. & S UZUKI , S. 2003. Mid to late Archaean (3.3–2.5 Ga) tonalitic crustal formation and high-grade metamorphism at Mt. Riiser-Larsen, Napier Complex, East Antarctica. Precambrian Research, 127, 215–228. I SHII , S., T SUNOGAE , T. & S ANTOSH , M. 2006. Ultrahigh-temperature metamorphism in the Achankovil Zone: implications for the correlation of crustal blocks in southern India. Gondwana Research, 10, 99–114. K ELSEY , D. E., W HITE , R. W., H OLLAND , T. J. B. & P OWELL , R. 2004. Calculated phase equilibria in K2O–FeO– MgO–Al2O3 –SiO2 –H2O for sapphirine – quartz-bearing mineral assemblages. Journal of Metamorphic Geology, 22, 559 –578. K RETZ , R. 1983. Symbols for rock-forming minerals. American Mineralogist, 68, 277– 279. L AMB , W., B ROWN , P. E. & V ALLEY , J. W. 1987. Postmetamorphic CO2-rich fluid inclusions in granulites. Contributions to Mineralogy and Petrology, 96, 485–495. L AMB , W., B ROWN , P. E. & V ALLEY , J. W. 1991. Fluid inclusions in Adirondack granulites: implications to the retrograde P– T path. Contributions to Mineralogy and Petrology, 107, 472–483. M OTOYOSHI , Y. & H ENSEN , B. J. 1989. Sapphirine– quartz–orthopyroxene symplectites after cordierite in the Archaean Napier Complex, Antarctica: Evidence for a counterclockwise P –T path? European Journal of Mineralogy, 1, 467– 471. M OTOYOSHI , Y. & H ENSEN , B. J. 2001. F-rich phlogopite stability in ultra-high-temperature metapelites from the Napier Complex, East Antarctica. American Mineralogist, 86, 1404– 1413.
UHT METAMORPHIC FLUIDS M OTOYOSHI , Y. & M ATSUEDA , H. 1984. Archaean granulites from Mt. Riiser-Larsen in Enderby Land, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 33, 103–125. N EWTON , R. C., S MITH , J. V. & W INDLEY , B. F. 1980. Carbonic metamorphism, granulites and crustal growth. Nature, 288, 45–50. O HYAMA , H., T SUNOGAE , T. & S ANTOSH , M. 2008. CO2-rich fluid inclusions in staurolite and associated minerals in a high-pressure and ultrahigh-temperature granulite from the Gondwana suture in southern India. Lithos, 101, 177– 190. O SANAI , Y., T OYOSHIMA , T., O WADA , M., T SUNOGAE , T., H OKADA , T. & C ROWE , W. A. 1999. Geology of ultrahigh-temperature metamorphic rocks from Tonagh Island in the Napier Complex, East Antarctica. Polar Geoscience, 12, 1– 28. O SANAI , Y., T OYOSHIMA , T., O WADA , M. ET AL . 2001a. Explanatory text of geological map of Tonagh Island, Enderby Land, Antarctica. Antarctic Geological Map Series, Sheet 38. National Institute of Polar Research, Tokyo. O SANAI , Y., T OYOSHIMA , T., O WADA , M., T SUNOGAE , T., H OKADA , T., C ROWE , W. A. & K USACHI , I. 2001b. Ultrahigh temperature sapphirine– osumilite and sapphirine– quartz granulites from Bunt Island in the Napier Complex, East Antarctica: reconnaissance estimation of P– T evolution. Polar Geoscience, 14, 1– 24. P ASTERIS , J. D. & C HOU , I.-M. 1998. Fluid-deposited graphitic inclusions in quartz: comparison between KTB (German Continental Deep-Drilling) core samples and artificially reequilibrated natural inclusions. Geochimica et Cosmochimica Acta, 62, 109–122. P ASTERIS , J. D., W OPENKA , B. & S EITZ , J. C. 1988. Practical aspects of quantitative laser Raman microprobe spectroscopy for the study of fluid inclusions. Geochimica et Cosmochimica Acta, 52, 979–988. R OEDDER , E. (ed.) 1984. Fluid Inclusions. Mineralogical Society of America, Reviews in Mineralogy, 12. S ANDIFORD , M. & P OWELL , R. 1986. Pyroxene exsolution in granulites from Fyfe Hills, Enderby Land, Antarctica: Evidence for 1000 8C metamorphic temperatures in Archean continental crust. American Mineralogist, 71, 946 –954. S ANTOSH , M. 1992. Carbonic fluids in granulites: cause or consequence? Journal of the Geological Society of India, 39, 375–399. S ANTOSH , M. & T SUNOGAE , T. 2003. Extremely high density pure CO2 fluid inclusions in a garnet granulite from southern India. Journal of Geology, 111, 1 –16. S ANTOSH , M., J ACKSON , D. H., H ARRIS , N. B. W. & M ATTEY , D. P. 1991. Carbonic fluid inclusions in South India granulites: evidence for entrapment during charnockite formation. Contributions to Mineralogy and Petrology, 108, 318–330. S ANTOSH , M., T SUNOGAE , T. & Y OSHIKURA , S. 2004. ‘Ultrahigh density’ carbonic fluids in ultrahigh-temperature crustal metamorphism. Journal of Mineralogical and Petrological Sciences, 99, 164–179. S ANTOSH , M., T SUNOGAE , T., L I , J. H. & L IU , S. J. 2007. Discovery of sapphirine-bearing Mg–Al granulites in the North China Craton: Implications for Paleoproterozoic
331
ultrahigh temperature metamorphism. Gondwana Research, 11, 263–285. S ARKAR , S., S ANTOSH , M., D ASGUPTA , S. & F UKUOKA , M. 2003. Very high density CO2 associated with ultrahigh-temperature metamorphism in the Eastern Ghats granulite belt, India. Geology, 34, 51– 54. S ATISH -K UMAR , M. 2005. Graphite-bearing CO2 –fluid inclusions in granulites: insight on graphite precipitation and carbon isotope evolution. Geochimica et Cosmochimica Acta, 69, 3841–3856. S HERATON , J. W., T INGEY , R. J., B LACK , L. P., O FFE , L. A. & E LLIS , D. J. 1987. Geology of an unusual Precambrian high-grade metamorphic terrane—Enderby Land and western Kemp Land, Antarctica. Bureau of Mineral Resources, Bulletin, 223. S HIMPO , M., T SUNOGAE , T. & S ANTOSH , M. 2006. First report of garnet–corundum rocks from Southern India: implications for prograde high-pressure (eclogitefacies?) metamorphism. Earth and Planetary Science Letters, 242, 111– 129. T ADOKORO , H., T SUNOGAE , T., S ANTOSH , M. & Y OSHIMURA , Y. 2007. First report of spinel þ quartz assemblage from Kodaikanal in the Madurai Block, southern India: implications for ultrahigh-temperature metamorphism. International Geology Review, 49, 1050– 1068. T ATEISHI , K., T SUNOGAE , T., S ANTOSH , M. & J ANARDHAN , A. S. 2004. First report of sapphirine þ quartz assemblage from southern India: implications for ultrahigh-temperature metamorphism. Gondwana Research, 7, 899–912. T OURET , J. L. R. 1977. The significance of fluid inclusions in metamorphic rocks. In: F RASER , D. G. (ed.) Thermodynamics in Geology. Reidel, Dordrecht, 203– 227. T OURET , J. L. R. 1985. Fluid regime in southern Norway, the record of fluid inclusions. In: T OBI , A. C. & T OURET , J. L. R. (eds) The Deep Proterozoic Crust in the North Atlantic Provinces. Reidel, Dordrecht, 517– 549. T OURET , J. L. R. 2001. Fluids in metamorphic rocks. Lithos, 55, 1 –25. T SUNOGAE , T. & VAN R EENEN , D. D. 2007. Carbonic fluid inclusions in sapphirine þ quartz bearing garnet granulite from the Limpopo Belt, southern Africa. Journal of Mineralogical and Petrological Sciences, 102, 57– 60. T SUNOGAE , T., S ANTOSH , M., O SANAI , Y., O WADA , M., T OYOSHIMA , T. & H OKADA , T. 2002. Very high-density carbonic fluid inclusions in sapphirinebearing granulites from Tonagh Island in the Archean Napier Complex, East Antarctica: implications for CO2 infiltration during ultrahigh-temperature (T . 1,100 8C) metamorphism. Contributions to Mineralogy and Petrology, 143, 279–299. T SUNOGAE , T., O SANAI , Y., O WADA , M., T OYOSHIMA , T., H OKADA , T. & C ROWE , W. A. 2003a. High fluorine pargasites in ultrahigh temperature granulites from Tonagh Island in the Archean Napier Complex, East Antarctica. Lithos, 70, 21– 38. T SUNOGAE , T., S ANTOSH , M., O SANAI , Y., O WADA , M., T OYOSHIMA , T., H OKADA , T. & C ROWE , W. A. 2003b. Fluid inclusions in an osumilite-bearing granulite from Bunt Island in the Archean Napier Complex,
332
T. TSUNOGAE ET AL.
East Antarctica: implications for a decompressional P –T path? Polar Geoscience, 16, 61– 75. T SUNOGAE , T., S ANTOSH , M. & D UBESSY , J. 2008. Fluid characteristics of high- to ultrahigh-temperature metamorphism in southern India: a quantitative Raman spectroscopic study. Precambrian Research, 162, 198–211.
V AN DEN K ERKHOF , A. M. & H EIN , U. F. 2001. Fluid inclusion petrography. Lithos, 55, 27– 47. W OPENKA , B. & P ASTERIS , J. D. 1986. Limitations to quantitative analysis of fluid inclusions in geological samples by laser Raman microprobe spectroscopy. Applied Spectroscopy, 40, 144– 151.
Origin of xenocrystic garnet and kyanite in clinopyroxene – hornblende-bearing adakitic meta-tonalites from Cape Hinode, Prince Olav Coast, East Antarctica YOSHIKUNI HIROI1, YOICHI MOTOYOSHI2, NAOTO ISHIKAWA3, TOMOKAZU HOKADA2 & KAZUYUKI SHIRAISHI2 1
Department of Earth Sciences, Graduate School of Science, Chiba University, 1-33, Yayoi-cho, Inage-ku, Chiba, 263-8522, Japan (e-mail:
[email protected])
2
National Institute of Polar Research, 1-9-10, Kaga, Itabashi-ku, Tokyo, 173-8522, Japan 3
Graduate School of Human and Environmental Studies, Kyoto University, Sakyo-ku, Kyoto, 606-8501, Japan Abstract: Xenocrystic garnet and kyanite, in addition to clinopyroxene and rare orthopyroxene, are newly found to occur in middle Proterozoic slightly metamorphosed adakitic trondhjemites and tonalites (meta-tonalites) at Cape Hinode on the eastern Prince Olav Coast in the latest Proterozoic –Early Palaeozoic Lu¨tzow-Holm Complex, East Antarctica. Textural and compositional features of garnet and kyanite suggest that these minerals formed most probably as restite phases of partial melting of mid-ocean ridge basalt (MORB) between 15 and 20 kbar pressure, and were entrained by the tonalitic magmas, which underwent fractional crystallization upon ascent to form cumulates that were also entrained and metamorphosed to basic– intermediate granulite blocks. Available geochronological data for the meta-tonalites indicate that all these events including MORB formation took place in the middle Proterozoic. The meta-tonalites and associated basic, calc-silicate, and pelitic rocks were emplaced as an allochthonous block in the Lu¨tzow-Holm Complex at the waning stage of its main regional metamorphism, most probably as a part of the final amalgamation of East and West Gondwana into the Gondwana supercontinent.
At Cape Hinode in the latest Proterozoic–Early Palaeozoic Lu¨tzow-Holm Complex (LHC) of East Antarctica occur slightly metamorphosed trondhjemites and tonalites (hereafter, collectively called meta-tonalites) associated with minor basic to intermediate, calc-silicate, and pelitic metamorphic rocks. Their major and trace element compositions closely resemble those of adakites and Archaean trondhjemites, tonalites and granodiorites (TTG), suggesting their young and hot slab-melt origin (Ikeda et al. 1997). This paper reports the first occurrence of xenocrystic garnet and kyanite in the meta-tonalites as the results of detailed field and petrographical re-examination, and briefly discusses their petrogenetic and tectonic significance.
Regional geological setting The LHC is one of the high-grade metamorphic terranes in the East Antarctic Shield (Fig. 1), and occurs along the Prince Harald, Soˆya, and Prince Olav coasts. Basement rocks are exposed in numerous ice-free areas along the coasts, separated by glaciers and the continental ice-sheet. To the east
of an inferred tectonic boundary lie the Proterozoic Rayner and Archaean Napier Complexes of Enderby Land, whereas to the SW, across several hundred kilometres of continental ice, lies the Yamato –Belgica Complex. Further to the west occur metamorphic and plutonic rocks of the Sør Rondane Mountains. The LHC is composed mainly of pelitic – psammitic and intermediate rocks with lesser amounts of basic –ultrabasic, calc-silicate, and siliceous rocks. These rocks are usually well layered, but migmatitic structures are also common. Some basic – ultrabasic rocks occur as isolated small blocks and lenses sitting within the pelitic – psammitic and intermediate rocks, essentially in the western part of the complex. They are most probably tectonically emplaced fragments of layered igneous complexes, and have experienced prograde and retrograde metamorphism in common with the host rocks (Hiroi et al. 1986, 1991; Suda et al. 2008). Less voluminous massive to foliated granites and pegmatites occur throughout the complex. Previous petrological work in the LHC has documented a systematic increase in metamorphic grade from east to west, ranging from upper
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 333 –350. DOI: 10.1144/SP308.16 0305-8719/08/$15.00 # The Geological Society of London 2008.
334
Y. HIROI ET AL.
Fig. 1. Map showing high-grade metamorphic terranes in the East Antarctic Shield. Cape Hinode is located on the eastern Prince Olav Coast in the latest Proterozoic–Early Palaeozoic Lu¨tzow-Holm Complex.
amphibolite facies on the eastern Prince Olav Coast to upper granulite facies around the head of Lu¨tzow-Holm Bay, then decreasing further west (i.e. Hiroi et al. 1983b, 1987, 1991; Shiraishi et al. 1984). Thus a thermal axis has been inferred around the head of Lu¨tzow-Holm Bay. However, metamorphic rocks occurring at Cape Hinode on the eastern Prince Olav Coast are exceptional, belonging to the granulite facies as revealed by Hiroi et al. (2006). At least two major deformation stages are recognized throughout the complex (Yoshida 1978; Ishikawa et al. 1994; Motoyoshi & Ishikawa 1997; Ikeda & Kawakami 2004; Kawakami & Ikeda 2004a, b). Ikeda & Kawakami (2004) argued that both deformation events may have occurred during the exhumation of the LHC while the rocks were still at high temperatures. Shiraishi et al. (1994, 2003) reported sensitive high-resolution ion microprobe (SHRIMP) U/Pb zircon ages of 520– 550 Ma for rocks of different metamorphic grade within the Lu¨tzow-Holm Complex and interpreted these zircon ages as the time of main regional metamorphism, making the LHC the youngest orogenic belt in the East Antarctic Shield. However, a meta-tonalite sample from Cape Hinode was anomalous; the youngest zircon age obtained was 1017 + 13 Ma.
Cape Hinode Cape Hinode is located on the eastern Prince Olav Coast at around 688090 S, 428400 E (Fig. 1). The
lithology over most of the area of about 20 km2 is dominated by relatively homogeneous, mediumto coarse-grained, dark grey to purple metatonalites, which are not known to occur in any other part of the LHC. Yanai & Ishikawa (1978) and Kanisawa et al. (1979) reported general petrography and major element bulk chemical compositions of rocks occurring in the Cape Hinode area. Yanai & Ishikawa (1978) called the metatonalites anorthositic gneisses because of their unusual mesoscopic appearance, which is similar to that of anorthosites. The meta-tonalites show only a weak foliation that is folded around a large-scale, tight antiform, trending NW –SE. Probable igneous layering and lamination are occasionally observed (Hiroi et al. 2006). Within the meta-tonalites occur small amounts of basic to intermediate and calc-silicate metamorphic rocks as isolated blocks. Basic metamorphic rocks are usually coarse-grained, massive to well-foliated, clinopyroxene or garnet amphibolite. They consist of hornblende, biotite, plagioclase and ilmenite, with clinopyroxene or garnet. Clinopyroxene and garnet commonly show partial replacement by hornblende + quartz and plagioclase þ hornblende þ biotite, respectively. Hiroi et al. (2006) found that some of these basic –intermediate rocks also contain orthopyroxene, which is visible to the naked eye as brown crystals intimately associated with light green clinopyroxene, green hornblende, black biotite, and grey plagioclase (Fig. 2). In thin section most of the orthopyroxene-bearing rocks (granulites) contain quartz, indicating that they are saturated
GARNET AND KYANITE IN ADAKITIC META-TONALITES
in SiO2. Neither olivine nor garnet has been found in the granulites, and the observed mineral assemblages are relatively simple, being composed of orthopyroxene, clinopyroxene, hornblende, biotite, sodic plagioclase, quartz, and magnetite. The granulites do not show any mineral textures indicative of prograde metamorphism. On the other hand, they often show such retrograde metamorphic features as the partial replacement of intergrown clino- and orthopyroxenes by calcic amphiboles with or without biotite, plagioclase, and quartz. Based on the modes of field occurrence and Fe–Mg-rich and Al-poor bulk-rock compositions, Hiroi et al. (2006) inferred that these basic – intermediate granulites originated from cumulates produced by fractional crystallization of the host tonalite magmas. Calc-silicate rocks are usually composed of grandite garnet, green salitic to esseneitic clinopyroxene, and calcic plagioclase with or without scapolite (Kanisawa et al. 1979; Kanisawa & Yanai 1978; Hiroi et al. 1987). Epidote is a
Fig. 2. (a) Coarse-grained, massive, kyanite–hornblendebearing meta-tonalite (upper part: sample YH04122003A) and basic–intermediate granulite block (lower part: sample YH04122003B). Plagioclase is dark grey, whereas quartz is milky white in the meta-tonalite. (b) Xenocrystic garnet occurring within hornblende–biotite meta-tonalite (sample YH04122002). The garnet is surrounded by its breakdown products composed of hornblende, biotite, plagioclase, and magnetite.
335
common and widespread retrograde mineral replacing garnet and plagioclase to varying degrees. A minor amount of basic to intermediate weakly metamorphosed dyke rocks is also known to occur within the meta-tonalites (Yanai & Ishikawa 1978; Suda et al. 2008). They are finegrained and massive, preserving original igneous textures well. They have amphibolite-facies mineral assemblages, but the degree of metamorphic recrystallization is distinctly different from that of the basic to intermediate blocks, suggesting that they were intruded into the pre-existing tonalites and accompanying rocks that already had experienced high-grade metamorphism. All these rocks are extensively intruded by pink pegmatites, which occasionally show graphic intergrowths of quartz and K-feldspar and contain blue beryl crystals, as reported by Yanai & Ishikawa (1978). Metatonalites and pegmatites are locally mylonitized. To the SW of Cape Hinode, separated from the meta-tonalites by a few hundred metres of intervening ice sheet, are several small outcrops of pelitic and psammitic gneisses. These consist of sillimanite –garnet– biotite gneiss and garnet– biotite gneiss with minor clinopyroxene amphibolite and garnet-bearing pegmatites, and are sheared to varying degrees. The sillimanite –garnet–biotite gneiss commonly contains retrograde kyanite that replaces pre-existing sillimanite to varying degrees, as reported by Hiroi et al. (1983a, 2006) and Motoyoshi et al. (2004, 2005). As mentioned above, Shiraishi et al. (1994, 2003) reported a SHRIMP U/Pb zircon age of 1017 + 13 Ma for a meta-tonalite sample from Cape Hinode. In addition, Shiraishi et al. (1995) reported a Sm –Nd whole-rock isochron age of 1031 + 69 Ma with an initial 143Nd/144Nd ratio of 0.5115520 + 0.000053 (MSWD ¼ 0.10, Nd ¼ þ 4.7 + 1.0) and a Rb-Sr whole rock isochron age of 1273 + 104 Ma with an initial 87 Sr/86Sr ratio of 0.70275 + 0.00017 (MSWD ¼ 7.4) for the meta-tonalites, and interpreted these ages as the times of igneous crystallization. Shiraishi et al. (1995) also calculated depleted mantle Nd model ages (TDM) for the meta-tonalites, using the model of DePaolo (1981, 1988). They range from 1070 to 1170 Ma and are essentially the same as, or slightly older than, the SHRIMP U –Pb zircon and Sm–Nd whole-rock isochron ages. Motoyoshi et al. (2004, 2005) reported similar but slightly younger monazite electron microprobe (EMP) ages (935 –1007 Ma) for sheared, retrograde kyanite-bearing pelitic gneiss and associated garnet–monazite-rich pegmatite from the southwestern part of the Cape Hinode area. They considered that the main sillimanitegrade regional metamorphic event and subsequent deformation and retrograde kyanite formation took
336
Table 1. Major and trace element analyses of meta-tonalites from Cape Hinode Type:
H
B
G
74010107
YH04122102
73123103
74010606
73123116
74010701
YH04122003A
74010115
74010105
73123106
74010113
wt% SiO2 TiO2 Al2O3 Fe2O3* MnO MgO CaO Na2O K2O P2O5 LOI Total
63.26 0.46 16.12 4.51 0.07 2.23 5.49 4.04 1.08 0.12 0.73 98.11
63.66 0.56 17.58 4.83 0.07 2.90 6.00 4.38 1.26 0.14
64.52 0.37 17.67 3.51 0.05 1.17 5.55 5.50 0.57 0.14 0.61 99.46
66.24 0.41 17.67 3.65 0.04 1.30 5.06 4.86 1.09 0.16 0.40 100.88
68.13 0.31 16.43 2.93 0.06 0.98 4.92 4.69 0.64 0.15 0.50 99.74
69.06 0.29 16.83 2.78 0.04 1.03 4.40 4.64 1.02 0.10 0.44 100.63
72.85 0.02 16.79 0.38 0.01 0.19 4.40 4.68 0.60 0.01
74.15 0.10 15.09 1.02 0.02 0.35 3.46 4.68 0.79 0.03 0.54 100.23
71.28 0.03 16.34 0.47 0.01 0.15 4.38 4.69 0.58 0.02 0.29 98.23
71.57 0.19 15.82 2.03 0.03 0.58 3.33 5.21 0.96 0.07 0.68 100.47
76.56 0.22 12.04 2.92 0.06 0.61 2.10 4.01 1.07 0.04 0.35 99.98
ppm Cr Ni V Zn Th Hf Y Zr
59.4 29 64 50 0.5 2.5 9 96
27.5 18 38 40 0.1 2.2 6 65
23.9 18 35 45 0.2 2.5 16 100
27.2 15 7 16 0.2 2.7 ,1 93
18.1 10 2 10 0.1 1.3 ,1 42
101.39 41 21 99 62 11 151
22.6 14 38 34 0.1 2.8 5 104
33.1 16 28 41 0.2 2.2 2 91
99.93 ,1 ,1 4 7 ,1 64
18.8 11 17 31 0.1 2 2 73
33.8 20 12 19 2.7 3.8 30 139
Y. HIROI ET AL.
Sample:
25.2 653 556 9.2 21 11 2.17 0.7 0.3 0.86 0.12
0.91 0.56 49.5 63.5 A, B
25 650 568
4 3 17 0.90 54.3 C
2.19 891 315 6.6 15 7 1.45 0.63 0.2 0.52 0.07
0.89 0.59 39.8 52.8 A, B
18 681 735 10.7 24 14 3.21 0.99 0.6 1.19 0.16
0.96 0.59 41.4 54.7 A, B
3.01 687 349 5.9 12 6 1.12 0.53 0.2 0.54 0.09
0.95 0.61 39.8 52.1 A, B
19.3 723 714 8.5 16 5 0.84 0.48 0.1 0.22 0.03
1.01 0.60 42.3 55.0 A, B
4 810 394
11 0 1 1.03 49.7 C
14.7 646 422 4.7 8 3 0.42 0.39 0.1 0.12 0.02
2.17 767 415 4.3 6 2 0.2 0.33 0.1 0.16 0.02
12.6 554 629 6.9 14 6 0.93 0.39 0.1 0.22 0.03
15.2 141 562 12.7 30 13 2.75 0.61 0.6 4.88 0.76
1.02 0.40 40.4 63.0 A, B
1.00 0.77 38.7 45.1 A, B
1.01 0.47 36.1 54.7 A, B
1.04 0.63 29.3 39.5 A, B
H, hornblende þ biotite + clinopyroxene; B, biotite; G, garnet þ biotite. ASI is Aluminium Saturation Index (mol. Al2O3/(CaO þ Na2O þ K2O)) of Zen (1986). ORR is Oxidation –Reduction Ratio (mol. FeO/(FeO þ Fe2O3)) by Kanisawa et al. (1979) (¼ data source B). 1 Mol. MgO/(MgO þ FeO (total Fe)). 2 Mol. MgO/(MgO þ FeO (recalculated based on ORR)). 3 Data source: A, Ikeda et al. (1997); B, Kanisawa et al. (1979); C, this study, by XRF at the NIPR. *Total iron given as Fe2O3.
GARNET AND KYANITE IN ADAKITIC META-TONALITES
Rb Sr Ba La Ce Nd Sm Eu Tb Yb Lu Cu Nb Co ASI ORR Mg-no. (a)1 Mg-no. (b)2 Data Source3
337
338
Y. HIROI ET AL.
place in the middle Proterozoic, because no data indicative of an Early Palaeozoic overprinting event were obtained. On the other hand, Fraser & McDougall (1995) reported K/Ar and 40Ar/39Ar mineral (biotite, muscovite, and hornblende) ages ranging from 480 + 5 Ma to 526 + 5 Ma for amphibolitic gneiss that occurred within meta-tonalites and pegmatites associated with highly migmatitic pelitic gneiss, indicating an Early Palaeozoic thermal overprint in common with the other bedrock exposures of the LHC. Thus, overgrowth of monazite and zircon did not take place, but resetting of K/Ar and 40Ar/39Ar mineral ages occurred during the last stage of Pan-African orogeny at Cape Hinode. Palaeomagnetic study of rocks from Cape Hinode by Ishikawa et al. (2006) also suggested that characteristic remanent magnetic components of the rocks are similar to those of the neighbouring LHC rocks and were acquired at around 500 Ma.
Meta-tonalites The meta-tonalites comprise mainly plagioclase and quartz with subordinate biotite or hornblende þ biotite. Minor K-feldspar, magnetite, ilmenite, zircon, apatite, monazite, titanite, and allanite are also present. Muscovite, chlorite, epidote, and carbonates are common retrograde minerals replacing hornblende, biotite and plagioclase. As Yanai & Ishikawa (1978) and Kanisawa et al. (1979) noted, the meta-tonalites are divided into two types: hornblende type (H-type) and biotite type
(B-type). B-type rocks sometimes contain a small amount of garnet. The presence of garnet has a great effect on the trace element bulk-rock chemistry, as will be mentioned below, and garnetbearing rocks are grouped separately as G-type. Major and trace element geochemical characteristics of the meta-tonalites were first discussed in some detail by Ikeda et al. (1997) on the basis of 10 whole-rock analyses as listed in Table 1, which also contains two new analyses. Except for a garnet-bearing sample, they are characterized by: (1) high Al2O3 (.15 wt% at 70 wt% SiO2); (2) high Sr (.400 ppm); (3) low Y (,16 ppm) and Yb (,1.2 ppm). On the other hand, garnet-bearing rocks are not only rich in Y and heavy REE (HREE) but also rich in Th, Hf, and Zr. In addition, they are poor in Sr. Kanisawa et al. (1979) pointed out that the rocks have relatively high Fe2O3/FeO ratios (Fe3þ/total Fe ¼ 40 –60 mol%) based on their wet chemical analyses of the same samples that Ikeda et al. (1997) studied. This is in agreement with the occurrence of magnetite and voluminous haematite lamellae in ilmenite. The following features can be additionally noted: (4) high SiO2 (.60 wt%); (5) low TiO2 (,0.6 wt%); (6) low MgO (,3 wt%) and low Mg-number (mol. Mg/(Mg þ Fe) ,50); (7) low Cr (,35 ppm) and Ni (,20 ppm) except for one sample. All these geochemical features are similar to those of high-SiO2 adakites of Martin et al. (2005) (Fig. 3a and b), Archaean TTG (Martin et al. 2005), and especially liquids obtained by high-pressure
Fig. 3. (a) Normative feldspar variation, (b) Sr– (K/Rb)– (SiO2/MgO) 100, and (c) Harker variation diagrams for the meta-tonalites from Cape Hinode. Basic–intermediate granulites (cumulates) are also plotted in the Harker variation diagrams. The compositions of liquids obtained by high-pressure (.10 kbar) partial melting experiments on meta-basaltic rocks are shown in the MgO v. SiO2 diagram. Experimental liquids are after Winther & Newton (1991), Rapp et al. (1991), Sen & Dunn (1994), Wolf & Wyllie (1994), Rapp (1995), Rapp & Watson (1995), Winther (1996), Springer & Seck (1997) and Skjerlie & Patin˜o Douce (2002).
GARNET AND KYANITE IN ADAKITIC META-TONALITES
339
Fig. 3. (Continued)
(.10 kbar) partial melting experiments on metabasaltic rocks by Rapp et al. (1991), Winther & Newton (1991), Sen & Dunn (1994), Wolf & Wyllie (1994), Rapp (1995), Rapp & Watson (1995), Winther (1996), Springer & Seck (1997), and Skjerlie & Patin˜o Douce (2002) (Fig. 3c). Hiroi et al. (2006) re-examined the Cape Hinode meta-tonalites in detail and found that
clinopyroxene and kyanite are not uncommon, though small in amount (Figs 4 and 5). Moreover, orthopyroxene rarely occurs together with clinopyroxene, hornblende and biotite in the meta-tonalites. Thus, orthopyroxene-bearing metatonalites have the same mineral assemblage as that of the most basic–intermediate granulites. Clinopyroxene and its breakdown products are found
340
Y. HIROI ET AL.
Fig. 4. (a) Photomicrograph of kyanite–clinopyroxene –hornblende-bearing meta-tonalite (sample HND-4). (b) Higher magnification of the area in the left-hand rectangle in (a). It should be noted clinopyroxene is often partially replaced or surrounded by hornblende. (c) Hornblende rimmed by fine-grained biotite.
GARNET AND KYANITE IN ADAKITIC META-TONALITES
341
Fig. 5. Photomicrographs of kyanite in meta-tonalites. (a, b) Higher magnification of the area in the right-hand rectangule in Figure 4a. It should be noted that kyanite is in contact with plagioclase and quartz. Hornblende accompanied by biotite occurs near the kyanite, but these minerals are never in direct contact. (c) Kyanite totally enclosed within plagioclase in sample YH04122003A. (d) Higher magnification of the area in the rectangle in (c). It should be noted that kyanite is fringed by spinel. Quartz is also enclosed within plagioclase.
only in H-type rocks, whereas kyanite occurs in both B-type and H-type rocks. Representative chemical compositions of minerals in kyanite– clinopyroxene-bearing H-type sample HND-4 are given in Table 2. Plagioclase is subhedral to anhedral and characteristically dark grey to purple in colour to the naked eye. In thin section, it is antiperthitic and contains numerous tiny inclusions of apatite, fluids, and unidentified opaque mineral(s) in addition to K-feldspar lamellae. In most rocks of both hornblende-bearing and hornblende-free types, plagioclase ranges from 30 to 36 mol% anorthite (an). Quartz is usually milky white in colour to the naked eye. It is also rich in fine inclusions of rutile, unidentified mineral(s) and fluids. Clinopyroxene is pale green and almost always shows textures of partial replacement by calcic amphiboles (hornblende or actinolite) sometimes intergrown with quartz, or by fine-grained mixtures of quartz and clay minerals (hereafter
referred to as pinite; Fig. 4). It is compositionally heterogeneous and poor in both Al (Al2O3 ,2.3 wt%) and Na (Na2O ,0.65 wt%). Hornblende also commonly shows partial replacement by finegrained biotite with or without quartz and epidote (Fig. 4). It is usually heterogeneous in composition and colour within a grain and from grain to grain, in addition to showing extensive and irregular exsolution of ilmenite. Biotite shows three distinct modes of occurrence: as large inclusion-free flakes, relatively large grains intergrown with many vermicular quartz grains, and fine aggregates replacing hornblende (Fig. 4). The large inclusion-free type grains are rich in TiO2, up to 5.2 wt%, as compared with the other types of grains. Small garnet grains found in some G-type meta-tonalites commonly show textures indicative of breakdown to biotite þ plagioclase or recrystallization. Magnetite usually occurs as grains intimately associated with ilmenite. Ilmenite always shows extensive exsolution of haematite lamellae (Fig. 6).
342
Y. HIROI ET AL.
Table 2. Representative EPM analyses of minerals in sample HND-4 Mineral: SiO2 TiO2 Al2O3 Cr2O3 FeO* MnO MgO CaO Na2O K2O Total O Si Ti Al Cr Fe3þ Fe2þ Mn Mg Ca Na K Total mg-no.
Ky
Cpx
Hbl-1
Hbl-2
Bt-1
Bt-2
Bt-3
Pl
Mag
Ilm
37.16 0.12 61.52 0.00 0.75 0.01 0.00 0.00 0.00 0.00 99.56
52.91 0.28 2.29 0.00 9.18 0.50 12.93 21.78 0.61 0.02 100.50
43.74 2.10 11.14 0.02 16.04 0.44 10.83 11.44 1.70 1.42 98.87
46.96 1.10 9.48 0.00 15.25 0.30 12.67 11.62 1.13 1.03 99.54
37.32 5.04 15.58 0.00 16.85 0.22 11.57 0.00 0.02 9.75 96.35
37.65 4.50 14.69 0.00 16.71 0.16 12.57 0.00 0.12 9.66 96.06
37.95 4.38 15.30 0.03 15.74 0.07 13.34 0.16 0.03 9.85 96.85
60.59 0.00 25.19 0.03 0.07 0.00 0.00 6.74 7.69 0.44 100.75
0.00 0.00 0.20 0.07 93.68 0.07 0.03 0.00 0.01 0.00 94.06
0.01 38.53 0.04 0.04 58.67 1.88 0.23 0.00 0.02 0.00 99.42
6 1.964 0.008 0.100 0.000
23 6.487 0.234 1.947 0.002
23 6.817 0.120 1.622 0.000
22 5.565 0.565 2.738 0.000
22 5.628 0.506 2.588 0.000
22 5.597 0.486 2.659 0.003
8 2.682 0.000 1.314 0.001
0.285 0.016 0.716 0.866 0.044 0.001 4.000
1.989 0.055 2.394 1.818 0.489 0.269 15.684
1.851 0.037 2.742 1.807 0.318 0.191 15.505
2.101 0.028 2.572 0.000 0.006 1.855 15.430
2.089 0.020 2.801 0.000 0.035 1.842 15.509
1.941 0.008 2.933 0.025 0.009 1.853 15.514
0.003 0.000 0.000 0.320 0.660 0.025 5.005
0.715
0.546
0.597
0.550
0.573
0.602
5 1.009 0.002 1.968 0.000 0.017 0.000 0.000 0.000 0.000 0.000 2.996
4 0.000 0.000 0.009 0.002 1.989 0.996 0.002 0.002 0.000 0.000 0.000 3.000
3 0.000 0.724 0.000 0.001 0.550 0.675 0.040 0.009 0.000 0.000 0.000 2.000
0.000
0.013
Hbl-1, core of large brownish green grain; Hbl-2, small grain fringing Cpx; Bt-1, large flake; Bt-2, small grain fringing Hbl; Bt-3, small grain intergrown with Cpx. *Total iron given as FeO.
Kyanite occurs not only in B-type meta-tonalites but also in (clinopyroxene-bearing) H-type metatonalites as anhedral grains up to 2 mm long, being always replaced by pinite partially from outside and along cracks (Figs 4 and 5). It is rarely fringed by fine-grained spinel (Fig. 5c and d). Although smaller grains are totally enclosed within large plagioclase crystals, larger grains are in direct contact with plagioclase and quartz (Fig. 5). However, kyanite is not in direct contact with clinopyroxene, hornblende, biotite or garnet. Kyanite does not show any textural sign of transformation from or to sillimanite as commonly observed in pelitic gneisses. The occurrence of kyanite in the meta-tonalites, especially clinopyroxene– hornblende-bearing ones, may be unusual and its origin will be discussed below. Xenocrystic garnet was found in the field as grains up to 3 cm in diameter; these are surrounded by symplectitic intergrowths of garnet breakdown products and occur within H-type meta-tonalite (Figs 2b and 6). Representative analyses of the xenocrystic garnet, the surrounding breakdown products, and the constituents of host H-type metatonalite (sample YH04122002) are given in
Table 3. In thin section, it is anhedral, being replaced from outside and along cracks by symplectitic intergrowths of calcic plagioclase (62– 65 an mol%), aluminous blue–green hornblende, biotite, magnetite, and rare scapolite (Fig. 6a). On the other hand, the garnet is compositionally homogeneous except for rims next to the replacing symplectitic intergrowths (Fig. 7), and contains inclusions of plagioclase, apatite, rutile, and ilmenite. Rutile is present not only as stout grains associated with plagioclase but also as numerous lamellae in garnet (Fig. 6b). It is noteworthy that plagioclase inclusions are almost always euhedral, and are relatively calcic (c. 58 an mol%) but extremely poor in K compared with the more calcic plagioclase in the symplectitic intergrowths. In addition, neither Mn enrichment nor Ca depletion is observed around such euhedral plagioclase inclusions within garnet (Fig. 7). Fe –Ti oxide assemblages are rutile þ ilmenite within garnet, magnetite in the replacing symplectitic intergrowths, and magnetite þ ilmenite in meta-tonalites. Ilmenite always shows extensive exsolution of haematite lamellae (Fig. 6c and d), regardless of their mode of occurrence.
GARNET AND KYANITE IN ADAKITIC META-TONALITES
343
Fig. 6. Photomicrographs of xenocrystic garnet, its breakdown products, and host hornblende–biotite meta-tonalite shown in Figure 2b. (a) Xenocrystic garnet is replaced from outside and along cracks by symplectitic intergrowths of calcic plagioclase, hornblende, biotite, magnetite, and rare scapolite. (b) Higher magnification of the area in the left-hand small rectangle in (a). Garnet contains inclusions of apatite, plagioclase, and ilmenite in addition to numerous rutile lamellae. Stout rutile grains are also present elsewhere within garnet. The included plagioclase is commonly euhedral. (c) Higher magnification of the area in the right-hand small rectangle in (a). Hornblende in the host metatonalite contains irregularly distributed ilmenite lamellae. (d) BSE image of haematite lamellae-rich ilmenite in (c).
344
Table 3. Representative EPM analyses of xenocrystic garnet, its breakdown products, and minerals in host meta-tonalite (sample YH04122002) Mineral:
O Si Ti Al Cr Fe3þ Fe2þ Mn Mg Ca Na K Total mg-no.
Garnet breakdown products
Host H-type meta-tonalite
Grt-core
Grt-rim
Pl*
Hbl
Bt
Pl
Scp
Mag
Hbl
Bt
Pl
Ilm
Hm
39.63 0.11 21.54 0.00 26.10 1.58 5.86 6.96 0.04 0.00 101.82
38.71 0.08 21.29 0.01 22.65 9.25 3.18 6.70 0.01 0.00 101.87
54.36 0.04 28.69 0.00 0.45 0.00 0.00 11.71 4.56 0.01 99.81
41.71 0.36 13.85 0.01 16.80 0.47 9.65 11.41 1.33 1.33 96.92
37.78 1.55 18.18 0.00 14.74 0.30 14.39 0.02 0.15 9.83 96.93
52.61 0.04 29.82 0.00 0.09 0.00 0.03 12.90 4.01 0.09 99.58
46.68 0.07 27.35 0.01 0.24 0.09 0.02 18.04 3.27 0.14 95.91
0.00 0.00 0.12 0.00 93.06 0.14 0.00 0.00 0.00 0.00 93.32
43.65 1.32 11.31 0.04 15.41 0.56 10.72 11.41 1.47 1.34 97.23
36.78 3.34 15.31 0.02 18.00 0.27 12.70 0.01 0.04 9.70 96.16
60.29 0.00 25.74 0.00 0.08 0.02 0.00 7.49 7.30 0.25 101.17
0.00 49.63 0.00 0.01 47.25 1.23 0.42 0.00 0.00 0.00 98.54
0.00 10.72 0.11 0.11 81.42 0.11 0.00 0.00 0.00 0.00 92.47
23 6.554 0.149 2.001 0.005
22 5.538 0.378 2.717 0.002
8 2.658 0.000 1.338 0.000
1.935 0.071 2.400 1.836 0.428 0.257 15.636
2.266 0.034 2.851 0.002 0.012 1.863 15.663
0.003 0.001 0.000 0.354 0.624 0.014 4.992
0.554
0.557
12 3.031 0.006 1.942 0.000
12 3.018 0.005 1.957 0.000
8 2.459 0.001 1.530 0.000
23 6.326 0.041 2.476 0.001
22 5.523 0.170 3.132 0.000
8 2.393 0.001 1.599 0.000
8 2.267 0.002 1.566 0.000
1.669 0.102 0.668 0.570 0.006 0.000 7.994
1.477 0.611 0.370 0.560 0.002 0.000 8.000
0.017 0.000 0.000 0.568 0.400 0.001 4.976
2.131 0.060 2.182 1.854 0.391 0.257 15.719
1.802 0.037 3.136 0.003 0.043 1.833 15.679
0.003 0.000 0.002 0.628 0.354 0.005 4.985
0.010 0.004 0.001 0.939 0.308 0.009 5.106
0.286
0.200
0.506
0.635
*Euhedral inclusion in garnet. † Total Fe as FeO.
4 0.000 0.000 0.007 0.000 1.995 0.995 0.005 0.000 0.000 0.000 0.000 4.000
3 0.000 0.951 0.000 0.000 0.098 0.908 0.027 0.016 0.000 0.000 0.000 2.000
3 0.000 0.211 0.003 0.002 1.573 0.208 0.002 0.000 0.000 0.000 0.000 2.000
0.015
0.000
Y. HIROI ET AL.
SiO2 TiO2 Al2O3 Cr2O3 FeO† MnO MgO CaO Na2O K2O Total
Xenocryst
GARNET AND KYANITE IN ADAKITIC META-TONALITES
345
Fig. 7. Colour elemental maps of xenocrystic garnet and its breakdown products in the area of the large rectangle in Figure 6a. The calcic inclusions in garnet are apatite and euhedral plagioclase, as shown in Figure 6b, and neither Ca depletion nor Mn enrichment haloes are observed around these inclusions.
346
Y. HIROI ET AL.
Discussion and conclusions Origin of xenocrystic garnet and kyanite and P –T evolution of meta-tonalites The meta-tonalites are chemically similar to the high-SiO2 adakites of Martin et al. (2005), Archaean TTG (Martin et al. 2005), and especially liquids obtained by high-pressure (.10 kbar) partial melting experiments of meta-basaltic rocks, as mentioned above. They are subdivided into H-type and B-type. H-type rocks have higher Mg-number than B-type rocks (Table 1). In addition, H-type rocks are generally poor in SiO2 compared with B-type rocks, and typically contain clinopyroxene or its breakdown products. The compositional range of meta-tonalites may be partially the result of fractional crystallization, as suggested by the continuous and coherent compositional trends on Harker variation diagrams (Fig. 3c) and by the occurrence of cumulates (basic –intermediate granulites; Hiroi et al. 2006; Fig. 3c). Both H-type and B-type rocks are subaluminous to slightly peraluminous (aluminium saturation index (ASI) of Zen (1986) ,1.04; Table 1), and are not expected to carry such aluminous minerals as garnet and kyanite at low pressures (,10 kbar). Therefore, the garnet and kyanite found in the meta-tonalites may be either xenocrysts entrained by tonalitic magmas from greater depth or retrograde phases locally formed at the expense of pre-existing minerals such as plagioclase. Not only the coarse-grained, texturally xenocrystic garnet in H-type meta-tonalite (sample YH04122002) but also small garnet grains in G-type meta-tonalites show definite signs of breakdown from outside and along cracks. The breakdown products of garnet contain calcic plagioclase associated with aluminous hornblende, biotite, and magnetite, which constitute a typical low-pressure amphibolite-facies mineral assemblage. Therefore, it can be concluded that garnet formed at high pressures and later underwent partial breakdown by decompression and reactions with surrounding magmas. The chemical composition of the garnet, especially c. 20 mol% grossular content, is similar to those produced by the high-pressure (.10 kbar) partial melting experiments on basaltic to tonalitic rocks by Wolf & Wyllie (1993, 1994), Patin˜o Douce & Beard (1995), Skjerlie & Johnston (1996), Skjerlie & Patin˜o Douce (2002) and Patin˜o Douce (2005). The occurrence of numerous rutile lamellae within garnet indicates that the garnet was originally rich in TiO2, as is the case with natural high-pressure garnets and those produced by the high-pressure (.10 kbar) experiments (e.g. Patin˜o Douce 2005). The occurrence of haematite lamellae-rich ilmenite inclusions within garnet
suggests a genetic relationship between the garnet and meta-tonalites, because the meta-tonalites commonly contain haematite lamellae-rich ilmenite. In addition, the occurrence of relatively calcic, euhedral plagioclase inclusions within garnet sets an upper pressure limit at around 20 kbar for the garnet formation, as revealed by the experimental study (Fig. 8). The microscopic mode of occurrence of kyanite suggests that it is not a retrograde phase locally replacing pre-existing plagioclase. Kyanite usually does not coexist with hornblende and clinopyroxene under amphibolite- and granulite-facies conditions because of their reaction relations, which will produce mineral assemblages including calcic plagioclase þ garnet, cummingtonite, and/ or orthopyroxene (Conrad et al. 1988; Rutter & Wyllie 1988; Carroll & Wyllie 1990; Beard & Lofgren 1991; Rusher 1991; Singh & Johannes
Fig. 8. P –T diagram showing middle Proterozoic P– T paths followed by meta-tonalites and pelitic gneisses at Cape Hinode. Calculated P –T conditions based on the Al-in-hornblende barometer of Anderson & Smith (1995) and the amphibole– plagioclase thermometer of Holland & Blundy (1994) for meta-tonalites are shown by the stippled area. Kyanite occurrence at high pressure is also shown on the basis of subsolidus and partial melting experiments on basaltic to tonalitic rocks by Lambert & Wyllie (1972, 1974), Stern et al. (1975), Stern & Wyllie (1978), Huang & Wyllie (1986), Winther (1996), Skjerlie & Patin˜o Douce (2002), and Patin˜o Douce (2005).
GARNET AND KYANITE IN ADAKITIC META-TONALITES
1996). However, they can be in equilibrium at high-pressure conditions such as the eclogite facies. Moreover, kyanite has been reported to occur as one of the run products of high-pressure (.15 kbar) subsolidus and partial melting experiments on basaltic to tonalitic rocks by Lambert & Wyllie (1972, 1974), Stern et al. (1975), Huang & Wyllie (1986), Winther (1996), Skjerlie & Patin˜o Douce (2002), and Patin˜o Douce (2005). Therefore, kyanite in the meta-tonalites may also be a restite phase and entrained by tonalitic magmas from greater depth, where tonalitic magmas may have formed by partial melting of basaltic sources (Fig. 8). During ascent of highpressure tonalitic magmas and restite phases of garnet and kyanite, these minerals would have reacted to produce low-pressure mineral assemblages including calcic plagioclase by the reaction grossular þ kyanite þ quartz ¼ anorthite: This may be the reason why garnet and kyanite are mutually excludive in the meta-tonalites. Calculated temperatures based on the amphibole– plagioclasethermometer of Holland & Blundy (1994) range from 640 to 800 8C because of the wide compositional variation of hornblende (Fig. 8). However, the hornblende–plagioclase pair in the breakdown products of xenocrystic garnet yields temperatures higher than 800 8C. Calculated temperatures based on two-pyroxene thermometers for basic – intermediate granulite blocks are also usually higher than 800 8C (Y. Hiroi, unpubl. data). Calculated pressures based on the Al-inhornblende barometer of Anderson & Smith (1995) range from 4 to 7 kbar (Fig. 8). These values may show the P –T conditions of the last stage of magmatic crystallization and subsequent metamorphic overprint. It is noteworthy that andalusite was formed at the waning stage of regional metamorphism of the LHC, as found in several bedrock exposures on the Prince Olav Coast (Hiroi et al. 1983a, 1995, unpubl. data).
Tectonic significance of the meta-tonalites The K/Ar and 40Ar/39Ar mineral ages determined by Fraser & McDougall (1995) for the rocks associated with meta-tonalites at Cape Hinode indicate that these rocks definitely experienced a Pan-African thermal overprint. In addition, the palaeomagnetic data of Ishikawa et al. (2006) for the rocks from Cape Hinode suggest that these rocks were situated close to the LHC by c. 500 Ma. However, the SHRIMP U/Pb zircon age for a meta-tonalite reported by Shiraishi et al. (1994) suggests that the main igneous and
347
subsequent metamorphic events took place in middle Proterozoic time. At the same time or slightly later, associated pelitic gneisses experienced sillimanite-grade peak metamorphism and subsequent deformation and retrograde recrystallization to form kyanite, as evidenced by the monazite EMP ages determined by Motoyoshi et al. (2004, 2005). The depleted mantle Nd model ages (TDM) for the meta-tonalites, ranging from 1.07 to 1.17 Ga, calculated by Shiraishi et al. (1995) are essentially the same as or slightly older than the SHRIMP U –Pb zircon and Sm–Nd whole-rock isochron ages. Therefore, an extremely short crustal residence time is suggested not only for the metatonalites but also for their source material. If the source material of the tonalite magmas was MORB, this is in good agreement with the model of high-SiO2 adakite and Archaean TTG formation by young and hot slab melting (e.g. Defant & Drummond 1990; Drummond & Defant 1990: Martin et al. 2005). The relatively high initial 143 Nd/144Nd ratio (Nd ¼ þ 4.7 +1.0) and low initial 87Sr/86Sr ratio for the meta-tonalite determined by Shiraishi et al. (1995) may be in agreement with the MORB source model (e.g. Peacock et al. 1994). In addition, Ikeda et al. (1997) and Ikeda & Shiraishi (1998) showed that on a La/Yb v. Yb diagram the analytical data show a good agreement with the calculated model of partial melting of eclogite of MORB composition. The main orogenic event of the LHC took place in the latest Proterozoic –Early Palaeozoic. Therefore, the middle Proterozoic rocks at Cape Hinode as a whole are considered to be allochthonous (Hiroi et al. 2006). Middle Proterozoic rocks occur in the neighbouring Rayner Complex (Fig. 1), but adakitic meta-tonalites have not been reported to occur there. On the other hand, similar adakitic meta-tonalites of middle Proterozoic age are known to occur in the southwestern part of the Sør Rondane Mountains (Asami et al. 1992; Ikeda & Shiraishi 1998; Fig. 1). If the Cape Hinode allochthonous block was derived from the Sør Rondane Mountains, we should construct a new tectonic model to transport and mix deep crustal rocks over a distance of c. 700 km. Moreover, the emplacement of the allochthonous block in the LHC must have taken place not before or at the peak of regional metamorphism of the LHC but during its waning and uplift stage. Thus, the latest Proterozoic –Early Palaeozoic orogenic event in this part of East Antarctica may have involved the mixing of rocks belonging to different complexes over a distance of 700 km, most probably as a part of the final amalgamation of East and West Gondwana into the Gondwana supercontinent.
348
Y. HIROI ET AL.
We would like to express our sincere thanks to T. Adachi and K. Seno of the National Institute of Polar Research for their kind help during EMP and XRF analyses of rocks. We are grateful to D. J. Ellis, N. Tsuchiya, and H. Ishizuka for their critical and fruitful comments. We also thank M. Satish-Kumar of Shizuoka University for his editorial contributions.
References A NDERSON , J. L. & S MITH , D. R. 1995. The effects of temperature and fO2 on the Al-in-hornblende barometer. American Mineralogist, 80, 549 –559. A SAMI , M., O SANAI , Z., S HIRAISHI , K. & M AKIMOTO , H. 1992. Metamorphic evolution of the Sør Rondane Mountains, East Antarctica. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Science. Terra, Tokyo, 7– 15. B EARD , J. S. & L OFGREN , G. E. 1991. Dehydration melting and water-saturated melting of basaltic and andesitic greenstones and amphibolites at 1, 3, and 6.9 kb. Journal of Petrology, 32, 365–401. C ARROLL , M. R. & W YLLIE , P. J. 1990. The system tonalite– H2O at 15 kbar and the genesis of calc-alkaline magmas. American Mineralogist, 75, 345– 357. C ONRAD , W. K., N ICHOLLS , I. A. & W ALL , V. J. 1988. Water-saturated and -undersaturated melting of metaaluminous and peraluminous crustal compositions at 10 kb: Evidence for the origin of silicic magmas in the Taupo Volcanic Zone, New Zealand, and other occurrences. Journal of Petrology, 29, 765– 803. D EFANT , M. J. & D RUMMOND , M. S. 1990. Derivation of some modern arc magmas by melting of young subducted lithosphere. Nature, 347, 662– 665. D E P AOLO , D. J. 1981. Neodymium isotopes in the Colorado Front Range and crust–mantle evolution in the Proterozoic. Nature, 291, 193–196. D E P AOLO , D. J. 1988. Neodymium Isotope Geochemistry. Springer, Berlin. D RUMMOND , M. S. & D EFANT , M. J. 1990. A model for trondhjemite –tonalite– dacite genesis and crustal growth via slab melting: Archean to modern comparisons. Journal of Geophysical Research, 95, 21503– 21521. F RASER , G. L. & M C D OUGALL , I. 1995. K/Ar and 40 Ar/39Ar mineral ages across the Lu¨tzow-Holm Complex, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 8, 137 –159. H IROI , Y., S HIRAISHI , K., Y ANAI , K. & K IZAKI , K. 1983a. Aluminum silicates in the Prince Olav and Soya Coasts, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 28, 115–131. H IROI , Y., S HIRAISHI , K., N AKAI , Y., K ANO , T. & Y OSHIKUKA , S. 1983b. Geology and petrology of Prince Olav Coasts, East Antarctica. In: O LIVER , R. L., J AMES , P. R. & J AGO , J. B. (eds) Antarctic Earth Science. Australian Academy of Science, Canberra, A.C.T., 32–35. H IROI , Y., S HIRAISHI , K., M OTOYOSHI , Y., K ANISAWA , S., Y ANAI , K. & K IZAKI , K. 1986. Mode of
occurrence, bulk chemical compositions, and mineral textures of ultramafic rocks in the Lu¨tzow-Holm Complex, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 43, 62–84. H IROI , Y., S HIRAISHI , K., M OTOYOSHI , Y. & K ATSUSHIMA , T. 1987. Progressive metamorphism of calc-silicate rocks from the Prince Olav and Soˆya Coasts, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 1, 73–97. H IROI , Y., S HIRAISHI , K. & M OTOYOSHI , Y. 1991. Late Proterozoic paired metamorphic complexes in East Antarctica, with special reference to the tectonic significance of ultramafic rocks. In: T HOMSON , M. R. A., C RAME , J. A. & T HOMSON , J. W. (eds) Geological Evolution of Antarctica. Cambridge, Cambridge University Press, 83–87. H IROI , Y., M OTOYOSHI , Y., S HIRAISHI , K. & E LLIS , D. J. 1995. The significance of euhedral calcic plagioclase inclusions in garnet from the Lu¨tzow-Holm Complex, East Antarctica: A textural indicator of partial melting in pelitic gneisses. Proceedings of NIPR Symposium on Antarctic Geosciences, 8, 107–120. H IROI , Y., M OTOYOSHI , Y., S ATISH -K UMAR , M., K AGASHIMA , S., S UDA , Y. & I SHIKAWA , N. 2006. Granulites from Cape Hinode in the amphibolitefacies eastern part of Prince Olav Coast, East Antarctica: New evidence for allochthonous block in the Lu¨tzow-Holm Complex. Polar Geoscience, 19, 89–108. H OLLAND , T. & B LUNDY , J. 1994. Non-ideal interactions in calcic amphiboles and their bearing on amphibole – plagioclase thermometry. Contributions to Mineralogy and Petrology, 116, 433–447. H UANG , W. L. & W YLLIE , P. J. 1986. Phase relationships of gabbro–tonalite– granite–water at 15 kbar with applications to differentiation and anatexis. American Mineralogist, 71, 301–316. I KEDA , T. & K AWAKAMI , T. 2004. Structural analysis of the Lu¨tzow-Holm Complex in Akarui Point, East Antarctica, and overview of the complex. Polar Geoscience, 17, 22–34. I KEDA , Y. & S HIRAISHI , K. 1998. Petrogenesis of the tonalitic rocks from the Sør Rondane Mountains, East Antarctica. Polar Geoscience, 11, 143– 153. I KEDA , Y., S HIRAISHI , K. & Y ANAI , K. 1997. Petrogenesis of the meta-trondhjemites from Cape Hinode, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 10, 102–110. I SHIKAWA , M., S HIRAISHI , K., M OTOYOSHI , Y., T SUCHIYA , N., S HIMURA , T. & Y ANAI , K. 1994. Geological map of Ongul Islands, Antarctica. Antarctic Geological Map Series, Sheet 36. National Institute of Polar Research, Tokyo. I SHIKAWA , N., M OTOYOSHI , Y., H IROI , Y., S ATISH K UMAR , M., K AGASHIMA , S. & S UDA , Y. 2006. Paleomagnetic study of the Cape-Hinode block in the Lu¨tzow-Holm Complex, East Antarctica. Abstracts of the Japan Geoscience Union Meeting 2006, E135– P001. K ANISAWA , S. & Y ANAI , K. 1982. Metamorphic rocks of the Cape Hinode district, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 21, 71–85.
GARNET AND KYANITE IN ADAKITIC META-TONALITES K ANISAWA , S., Y ANAI , K. & I SHIKAWA , K. 1979. Major element chemistry of metamorphic rocks of the Cape Hinode district, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 14, 164–171. K AWAKAMI , T. & I KEDA , T. 2004a. Timing of ductile deformation and peak metamorphism in Skallevikhalsen, Lu¨tzow-Holm Complex, East Antarctica. Polar Geoscience, 17, 1– 11. K AWAKAMI , T. & I KEDA , T. 2004b. Structural evolution of Ongul Islands, Lu¨tzow-Holm Complex, East Antarctica. Polar Geoscience, 17, 12– 21. L AMBERT , I. B. & W YLLIE , P. J. 1972. Melting of gabbro (quartz eclogite) with excess water to 35 kilobars, with geological applications. Journal of Geology, 80, 693–708. L AMBERT , I. B. & W YLLIE , P. J. 1974. Melting of tonalite and crystallization of andesite liquid with excess water to 30 kilobars. Journal of Geology, 82, 88– 97. M ARTIN , H., S MITHIES , R. H., R APP , R., M OYEN , J.-F. & C HAMPION , D. 2005. An overview of adakite, tonalite –trondhjemite –granodiorite (TTG), and sanukitoid: relationships and some implications for crustal evolution. Lithos, 79, 1– 24. M OTOYOSHI , Y. & I SHIKAWA , M. 1997. Metamorphic and structural evolution of granulites from Rundva˚gshetta, Lu¨tzow-Holm Bay, East Antarctica. In: R ICCI , C. A. (ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antartica, Siena, 65– 72. M OTOYOSHI , Y., H OKADA , T., H IROI , Y. & S HIRAISHI , K. 2004. EMP dating on Cape Hinode of the Lu¨tzowHolm Complex, East Antarctica. Abstracts of 24th Symposium on Antarctic Geoscience, 23. M OTOYOSHI , Y., H IROI , Y., H OKADA , T. & S HIRAISHI , K. 2005. Geological belonging of Cape Hinode metamorphics, Lu¨tzow-Holm Complex, East Antarctica. Abstracts Volume of 2005 Annual Meeting of the Japanese Association of Mineralogists, Petrologists and Economic Geologists, 224. P ATIN˜ O D OUCE , A. E. 2005. Vapor-absent melting of tonalite at 15–32 kbar. Journal of Petrology, 46, 275–290. P ATIN˜ O D OUCE , A. E. & B EARD , J. S. 1995. Dehydration-melting of biotite gneiss and quartz amphibolite from 3 to 15 kbar. Journal of Petrology, 36, 707–738. P EACOCK , S. M., R USHMER , T. & T HOMPSON , A. B. 1994. Partial melting of subducting oceanic crust. Earth and Planetary Science Letters, 121, 227–244. R APP , R. P. 1995. Amphibole-out phase boundary in partially melted metabasalt, its control over liquid fraction and composition, and source permeability. Journal of Geophysical Research, 100, 15601–15610. R APP , R. P. & W ATSON , E. B. 1995. Dehydration melting of metabasalt at 8 –32 kbar: Implications for continental growth and crust–mantle recycling. Journal of Petrology, 36, 891– 931. R APP , R. P., W ATSON , E. B. & M ILLER , C. F. 1991. Partial melting of amphibolites/eclogite and the origin of Archean trondhjemites and tonalites. Precambrian Research, 51, 1– 25.
349
R USHER , T. 1991. Partial melting of two amphibolites: contrasting experimental results under fluid-absent conditions. Contributions to Mineralogy and Petrology, 107, 41– 59. R UTTER , M. J. & W YLLIE , P. J. 1988. Melting of vaporabsent tonalite at 10 kbar to simulate dehydration melting in the deep crust. Nature, 331, 159–160. S EN , C. & D UNN , T. 1994. Dehydration melting of a basaltic composition amphibolite at 1.5 and 2.0 GPa: implication for the origin of adakites. Contributions to Mineralogy and Petrology, 117, 394–409. S HIRAISHI , K., H IROI , Y. & O NUKI , H. 1984. Orthopyroxene-bearing rocks from the Tenmondai and Naga-iwa Rocks in the Prince Olav Coast, East Antarctica; first appearance of orthopyroxene in progressive metamorphic sequence. Memoirs of National Institute of Polar Research, Special Issue, 33, 126– 144. S HIRAISHI , K., E LLIS , D. J., H IROI , Y., F ANNING , Y., M OTOYOSHI , Y. & N AKAI , Y. 1994. Cambrian orogenic belt in East Antarctica and Sri Lanka: implications for Gondwana assembly. Journal of Geology, 102, 47– 65. S HIRAISHI , K., K AGAMI , H. & Y ANAI , K. 1995. Sm– Nd and Rb– Sr isochron ages for meta-trondhjemites from Cape Hinode. Proceedings of NIPR Symposium on Antarctic Geosciences, 8, 130–136. S HIRAISHI , K., H OKADA , T., F ANNING , C. M., M ISAWA , K. & M OTOYOSHI , Y. 2003. Timing of thermal events in eastern Dronning Maud Land, East Antarctica. Polar Geoscience, 16, 76–99. S INGH , J. & J OHANNES , W. 1996. Dehydration melting of tonalites. Part II. Composition of melts and solids. Contributions to Mineralogy and Petrology, 125, 26–44. S KJERLIE , K. P. & J OHNSTON , A. D. 1996. Vapor-absent melting from 10 to 20 kbar of crustal rocks that contain multiple hydrous phases: Implications for anatexis in the deep to very deep continental crust and active continental margins. Journal of Petrology, 37, 661– 691. S KJERLIE , K. P. & P ATIN˜ O D OUCE , A. E. 2002. The fluid-absent partial melting of a zoisite-bearing quartz eclogite from 1.0 to 3.2 GPa; Implications for melting in thickened continental crust and for subduction-zone processes. Journal of Petrology, 43, 291– 314. S PRINGER , W. & S ECK , H. A. 1997. Partial fusion of basic granulites at 5 to 15 kbar: implications for the origin of TTG magmas. Contributions to Mineralogy and Petrology, 127, 30–45. S TERN , C. R. & W YLLIE , P. J. 1978. Phase compositions through crystallization intervals in basalt-andesiteH2O at 30 kbar with implications for subduction zone magmas. American Mineralogist, 63, 641– 663. S TERN , C. R., H UANG , W. L. & W YLLIE , P. J. 1975. Basalt-andesite-rhyolite-H2O: crystallization intervals with excess H2O and H2O-undersaturated liquidus surfaces to 35 kilobars, with implications for magma genesis. Earth and Planetary Science Letters, 28, 189– 196. S UDA , Y., K AWANO , Y., Y AXLEY , G., K ORENAGA , H. & H IROI , Y. 2008. Magmatic evolution and tectonic setting of metabasites from Lu¨tzow-Holm
350
Y. HIROI ET AL.
Complex, East Antarctica. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctic: a Key to the East – West Gondwana Evolution. Geological Society, London, Special Publications, 308, 211 – 233. W INTHER , K. T. 1996. An experimentally based model for the origin of tonalitic and trondhjemitic melts. Chemical Geology, 127, 43–59. W INTHER , K. T. & N EWTON , R. C. 1991. Experimental melting of an hydrous low-K tholeiite: evidence on the origin of Archaean cratons. Bulletin of Geological Society of Denmark, 39, 213 –228. W OLF , M. B. & W YLLIE , P. J. 1993. Garnet growth during amphibolite anatexis: Implications of a garnetiferous restite. Journal of Geology, 101, 357–373.
W OLF , M. B. & W YLLIE , P. J. 1994. Dehydration-melting of amphibolite at 10 kbar: the effects of temperature and time. Contributions to Mineralogy and Petrology, 115, 369– 383. Y ANAI , K. & I SHIKAWA , T. 1978. Geological map of Cape Hinode, Antarctica. Antarctic Geological Map Series, Sheet 11. National Institute of Polar Research, Tokyo. Y OSHIDA , M. 1978. Tectonics and petrology of charnockites around Lu¨tzow-Holmbukta, East Antarctica. Journal of Geosciences, Osaka City University, 21, 65–152. Z EN , EA . 1986. Aluminum enrichment in silicate melts by fractional crystallization: Some mineralogic and petrographic constraints. Journal of Petrology, 27, 1095– 1117.
Kornerupine sensu stricto associated with mafic and ultramafic rocks in the Lu¨tzow-Holm Complex at Akarui Point, East Antarctica: what is the source of boron? T. KAWAKAMI1, E. S. GREW2, Y. MOTOYOSHI3, C. K. SHEARER4, T. IKEDA5, P. V. BURGER4 & I. KUSACHI6 1
Department of Geology and Mineralogy, Graduate School of Science, Kyoto University, Kyoto 606-8502, Japan, (e-mail:
[email protected]) 2
Department of Earth Sciences, University of Maine, Bryand Global Sciences Center, Orono, ME 04469-5790, USA
3
National Institute of Polar Research, 1-9-10 Kaga, Itabashi-ku, Tokyo 173-8515, Japan
4
Department of Earth and Planetary Sciences, University of New Mexico, Albuquerque, NM 87131-0001, USA
5
Department of Earth and Planetary Sciences, Graduate School of Science, Kyushu University, 33 Hakozaki, Fukuoka 812-8581, Japan 6
Department of Earth Sciences, Faculty of Education, Okayama University, 3-1-1 Tsushima-naka, Okayama 700-8530, Japan Abstract: Kornerupine, (A, Mg, Fe)(Al, Mg, Fe)9(Si, Al, B)5O21(OH, F), is known from only five mafic or ultramafic settings worldwide (of the .70 localities overall). We report a sixth occurrence from Akarui Point in the Lu¨tzow-Holm Complex, East Antarctica, where two ruby corundum (0.22–0.34 wt% Cr2O3)–plagioclase lenses are found at the same structural level as boudinaged ultrabasic rocks in hornblende gneiss and amphibolite. Ion microprobe analyses of kornerupine give 13–59 ppm Be, 181– 302 ppm Li, and 5466– 6812 ppm B, corresponding to 0.38– 0.47 B per 21.5 O; associated sapphirine also contains B (588– 889 ppm). Peak metamorphic conditions are estimated to be 770–790 8C and 7.7– 9.8 kbar. Kornerupine encloses tourmaline and plagioclase, which suggests the prograde reaction tourmaline (1) þ plagioclase (.An34) þ sapphirine + spinel ! kornerupine þ corundum (ruby) þ plagioclase (,An82) + (fluid or melt). Alternatively, kornerupine and tourmaline could have formed sequentially under nearly constant P– T conditions during the infiltration of fluid that was originally B-bearing, but then progressively lost Na (or gained Ca) and B through reaction with mafic rocks. Kornerupine later reacted with H2O–CO2 fluid in cracks at P– T conditions in the andalusite stability field: kornerupine þ plagioclase þ (Na, K, + Si in fluid) ! tourmaline þ biotite þ corundum (sapphire) + magnesite + andalusite þ (Ca in fluid). Secondary tourmaline differs from the included tourmaline in containing less Ti and having a higher Na/(Na þ Ca þ K) ratio. There are two possible scenarios for introducing B into the lenses: (1) infiltration of boron-bearing aqueous fluids released by prograde breakdown of muscovite in associated metasedimentary rocks; (2) hydrothermal alteration of mafic and ultramafic rocks by seawater prior to peak metamorphism. The latter scenario is consistent with an earlier suggestion that Akarui Point could be part of an ophiolite complex developed between the Yamato–Belgica and Rayner complexes.
Compared with other petrological reservoirs in the Earth’s crust, boron is characteristically most abundant in sediments and sedimentary rocks and least abundant in mafic and ultramafic rocks (e.g. Leeman & Sisson 1996). Processes associated with metamorphism lead to depletion of boron, and the extent of depletion increases with the grade of metamorphism. Thus, the formation of
borosilicate minerals of the tourmaline group and kornerupine series in mafic and ultramafic rocks metamorphosed in the upper amphibolite and granulite facies is particularly surprising. In the present paper, we report kornerupine associated with hornblende gneiss and amphibolite in the transition zone between the amphibolite and granulite facies at Akarui Point, Prince Olav Coast, East
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 351 –375. DOI: 10.1144/SP308.17 0305-8719/08/$15.00 # The Geological Society of London 2008.
352
T. KAWAKAMI ET AL.
Fig. 1. Geological map of Akarui Point after Yanai et al. (1984) showing location of the kornerupine-bearing lenses. The six named exposures in the inset map belong to the Lu¨tzow-Holm Complex; the bold line marked with query is the inferred boundary between this complex to the west and the Rayner Complex to the east.
Antarctica (Fig. 1) for the first time. Akarui Point is only the sixth reported occurrence worldwide of a kornerupine series mineral associated with mafic and ultramafic rocks (Table 1). In general, the kornerupine series, kornerupine sensu stricto and its boron-dominant analogue prismatine, both (A, Mg, Fe)(Al, Mg, Fe)9(Si, Al, B)5O21(OH, F), have been reported from over 70 localities worldwide in upper amphibolite- and granulite-facies rocks, including three others in the Precambrian shield of East Antarctica (Grew 1996): Windmill Islands (e.g. Post et al. 1997), Larsemann Hills (e.g. Carson et al. 1995), and Øygarden Islands (Kelly & Harley 2004). The present paper is a detailed description of the two kornerupine-bearing lenses and a discussion of possible scenarios for the origin and retention of boron in a mafic –ultramafic setting. The source of boron for kornerupine formation in these mafic and ultramafic associations and its retention at higher grades of metamorphism are even more
problematic than in the case of the more widespread association of kornerupine and prismatine with pelitic rocks (e.g. Grew 1996), because boron is overall less abundant in mafic and ultramafic rocks (Leeman & Sisson 1996). Mineral abbreviations are after Kretz (1983).
Metamorphism of the Lu¨tzow-Holm Complex The Lu¨tzow-Holm Complex is a Cambrian orogenic belt bounded by the Late Proterozoic Rayner Complex to the east and by the Late Proterozoic to Early Palaeozoic Yamato –Belgica Complex to the west (Shiraishi et al. 1992; Fig. 1). Metamorphic grade increases progressively from upper amphibolite facies on the Prince Olav Coast to granulite facies in Lu¨tzow-Holm Bay (Hiroi et al. 1991). The transition between the two facies is characterized by the presence of orthopyroxene or gedrite in
Table 1. Occurences of kornerupine in association with mafic and ultramafic rocks Locality
Occurrence
Origin (1) Synmetamorphic introduction of potassic aqueous fluid; (2) premetamorphic hydrothermal alteration by seawater Contamination by metasediments through which magma passed Metamorphism after hydration of mafic and ultramafic igneous rocks Contact metamorphism and metasomatism around lherzolite Hydration of Grt – Spl assemblage by fluids of mantle or crustal origin (l) Synmetamorphic introduction of fluid; (2) premetamorphic hydrothermal alteration by seawater
Fiskenæsset anorthosite complex, SW Greenland
Lenses along upper contact of anorthosite with country rock
An, Spr, Hbl, Ged, Phl, Crn, Crd; 2nd Chl (type locality); 2nd Tur (other)
Moon Mountain, Adirondacks, New York, USA
Ultramafic lens in layered anorthosite sill
Hbl, Spl, Spr, Phl, Opx, Ged; 2nd Tur
Kittila¨, Finnish Lapland
Hornblende gneiss, amphibolite and hornblendite Contact zone between peridodite (lherzolite) and marble Garnet –spinel xemolith in kimberlite
Hbl, An, Crn, chromite; 2nd Tur, Czo, Chl, Ms
Lenses in hornblende gneiss and amphibolite
Pl, Crn, Bt, Tur, Spr, Spl, 2nd Tur
Lherz peridotite, Pyrenees, France Moses Rock, Utah, USA Akarui Point
Ath, Spr, Hbl, Spl, Tur, Rt; 2nd Cal, vermiculite (from Phl?), saponite (from Crd?) Grt, Spl, Chl, Crn
Sources Herd et al. (1969); Herd (1973); Grew et al. (1987); Peck & Valley (1996)
Farrer & Babcock (1993); Farrar (1995); Farrar & Miller (2005) Haapala et al. (1971); Schumacher (1984) Monchoux (1972); Vielzeuf & Kornprobst(1984) Padovani & Tracy (1981); Smith (1995) This paper
¨ TZOW-HOLM COMPLEX KORNERUPINE FROM THE LU
Main associated minerals
An, anorthite; Spy, sapphirine; Hbl, hornblende; Ged, gedrite; Phl, phlogopite; Crn, corundum; Crd, cordierite; Chl, chlorite; Tur, tourmaline; Spl, spinel; Opx, orthopyroxene; Czo, clinozoisite; Ms, muscovite; Ath, anthophyllite; Rt, rutile; Cal, calcite; Grt, garnet.
353
354
T. KAWAKAMI ET AL.
a few gneisses (Shiraishi et al. 1984), and the peak metamorphic conditions for the transitional zone were determined to be 7.2 –7.5 kbar, 750 8C for Tenmondai Rock, 12 km east of Akarui Point (Hiroi et al. 1983a). Ultrahigh-temperature (UHT) metamorphic conditions of c. 1000 8C and c. 11 kbar were attained in the thermal axis near Rundva˚gshetta in the southern Lu¨tzow-Holm Bay (Motoyoshi 1986; Kawasaki et al. 1993; Motoyoshi & Ishikawa 1997). The presence of relict prograde kyanite and staurolite included in garnet and plagioclase (Hiroi et al. 1983b; Motoyoshi 1986) and reaction textures such as orthopyroxene þ plagioclase þ spinel intergrowth replacing garnet in ultramafic rocks (Hiroi et al. 1986) are evidence for a ‘clockwise’ P–T path for the complex. Prograde breakdown of staurolite itself to a sillimanite-bearing assemblage in the Prince Olav Coast, but to a kyanite-bearing one in Lu¨tzow-Holm Bay, suggests regional variations in pressure (Hiroi et al. 1983b, 1987). Timing of the peak regional metamorphism is estimated by sensitive highresolution ion microprobe (SHRIMP) dating to be between 521 + 9 and 553 + 6 Ma (Shiraishi et al. 1992, 1994). It is also considered that period between 520 and 500 Ma involved c. 600 8C of cooling, and c. 23 km of relatively rapid exhumation in Rundva˚gshetta (Fraser et al. 2000).
biotite-bearing hornblende gneiss and hornblendebearing biotite gneiss are locally migmatitic and grade into one another. Interboudin spaces are commonly filled with leucosome that could represent partial melts of local origin (e.g. Ikeda & Kawakami 2004). Relict kyanite occurs as inclusions in garnet and plagioclase in the garnet–biotite gneiss, indicating that the regional metamorphism was of the kyanite–sillimanite type (Hiroi et al. 1983b). Orthopyroxene is one of the main constituent minerals in the ultrabasic granulite, pyroxene gneiss and twopyroxene-bearing metabasite (Yanai et al. 1984). Granite and pegmatite veins with pinkish Kfeldspar are discordant to the penetrative foliation (Yanai et al. 1984). Three ductile deformational stages have been deduced from fold interference patterns (e.g. Ikeda & Kawakami 2004, and references therein). The foliation (Sm21) subparallel to the compositional layering strikes NW–SE and dips steeply towards NE or SW, and is deformed by Fm folds, one of which is the kilometre-scale synform on Sansyoku terrace where kornerupine was found (Fig. 1). Boudinaged ultrabasic rocks (Hiroi et al. 1986) and pyroxenites are folded by this synform (Ikeda & Kawakami 2004). Local upright, open folds (Fmþ1) are the later generation.
Borosilicates in the Lu¨tzow-Holm Complex
Field survey and analytical methods
Other than kornerupine at Akarui Point, no prograde borosilicates have been reported in the Lu¨tzowHolm Complex. Post-metamorphic tourmaline has been found in Lu¨tzow-Holm Bay as (1) a secondary mineral filling cracks in garnet or replacing garnet in corundum–spinel–garnet–sillimanite leucogneiss (Skallevikshalsen, T. Kawakami, unpubl. data) or (2) a constituent of the quartz þ calcite vein crosscutting the penetrative foliation (Skallevikshalsen, T. Kawakami, unpubl. data), and (3) a constituent of pegmatite veins (Rundva˚gshetta, Skallevikshalsen, Skarvsnes and Skallen; Osanai et al. 2004; E. S. Grew, unpubl. data). The source of boron for these tourmaline occurrences is not known.
Geological sketch of Akarui Point Akarui Point is one of the exposures in the transitional zone from amphibolite facies to granulite facies on the Prince Olav Coast (Fig. 1, Hiroi et al. 1983b, 1986; Yanai et al. 1984). Garnet–biotite, biotite– hornblende and hornblende–biotite gneisses are dominant, with subordinate ultrabasic rocks, pyroxene gneiss and two-pyroxene-bearing metabasite as boudins mainly in the biotite–hornblende gneiss. The
Y. Motoyoshi made the original discovery of one of the kornerupine-bearing lenses studied in this paper during the 42nd Japanese Antarctic Research Expedition (JARE-42, 2000–2002), and T. Ikeda, T. Kawasaki, Y. Kawano and T. Kawakami studied two kornerupine-bearing lenses in detail during the summer season of JARE-44 (2002–2003). Electron microprobe analyses of constituents other than Li, Be and B were obtained on three samples of kornerupine-bearing rock (AKR2002, TK2002122104 and I-010), hornblende gneiss I-009 and pyroxenite I-011 with a JEOL JXA8900R electron microprobe at the National Institute of Advanced Industrial Science and Technology (AIST) and Okayama University of Science (OUS), and with a JEOL6400 equipped with an Oxford Instruments light EDS detector and Link ISIS SEM quantitative software at The Australian National University (ANU). Analytical conditions at AIST and OUS using wavelength-dispersive spectroscopy are 15 kV accelerating voltage, 12 nA beam current and 1 mm beam diameter, and those at ANU are 15 kV accelerating voltage, 1 nA sample current and 3– 5 mm beam diameter. Natural and synthetic silicates, oxides and metals were used for standards and ZAF correction was applied for the analyses. Presence of boron is also
¨ TZOW-HOLM COMPLEX KORNERUPINE FROM THE LU
considered in the ZAF correction of kornerupine, and presence of 4 wt% OH is assumed for the ZAF correction of biotite in sample AKR2002A used for secondary ion mass spectrometry (SIMS) analyses. For the ZAF correction of biotite in other samples, presence of OH was not considered. Including OH in the ZAF correction has some effect on the calculated oxide wt%, but the effect on calculated formulae was confirmed to be negligible. X-ray elemental mapping of garnets was carried out with a JEOL JXA-8800R superprobe at AIST. Kornerupine, sapphirine, biotite and plagioclase were analysed for Li, Be and B in two gold-coated polished thin sections 2.5 cm in diameter (A and B of sample AKR2002) by SIMS on a Cameca ims 4f ion microprobe at the Institute of Meteoritics at the University of New Mexico (UNM). Analyses were made using a primary beam of 16O – ions accelerated through a nominal potential of 12.5 kV, and a primary beam current of 8 nA, with a spot size of 8–12 mm. Sputtered secondary ions were energyfiltered using a sample offset voltage of 75 V, and an energy window of +25 V to reduce isobaric interferences. For Li and B a mass resolution of c. 320 (routine operating conditions) was sufficient, but moderate mass resolution (c. 1000– 1100) was required to separate the 9Beþ from 27Al3þ; consequently, Be was analysed in a separate session. Counting times were 5 s (7Liþ), 4 s (9Beþ), 5 s (11Bþ), and 4 s (30Siþ) while measuring Li and B, and 2 s (30Siþ) while measuring Be. Each mass was analysed for 10 cycles. Measured intensities were normalized for ion yield by dividing the measured intensity of a given mass by the measured intensity of 30Siþ, and then multiplying by the SiO2 concentration for the sample determined by JEOL JXA-8900 at AIST. Absolute concentration was determined using calibration curves defined by the known concentration v. measured, Si-normalized intensity values for Li, Be and B in our standards. We attempted to match as closely as possible the matrix of the standards and unknowns. In analysing Li and B, we used prismatine and tourmaline standards from Dyar et al. (2001) and Cooper (1997) for kornerupine and sapphirine, glass standards from NIST and J. Evensen (e.g. Evensen et al. 1999) for plagioclase, and these glass standards together with biotite standards from the UNM laboratory and a prismatine standard for biotite. In analysing Be, we used surinamite and khmaralite (Grew et al. 2006) for kornerupine, sapphirine and biotite, and glass standards for plagioclase. We took measures to avoid B contamination of the thin-section surface, which were carbon coated for electron microprobe analysis (EMPA) only after completion of the SIMS analyses. Prior to gold coating and the SIMS analyses, samples were
355
washed with ultrapure, deionized water and soaked in 1% mannitol solution. The mannitol solution extracts B from the surface; it should also remove any polishing powder, a potential source of B contamination. Particles of polishing powder lodged in cracks would not have been removed by this cleaning, but in our analyses we tried to stay away from cracks. Surface contamination is further limited because the analytical area is generally rastered prior to collecting data. Rastering cleans off surface contaminants and allows the analyst to image an area with the ion beam to check that the spot is suitable for analysis. During the analysis we check for surface contamination by observing changes in B with depth of beam penetration. Grew et al. (2006) reported that repeated calibrations of the three Evensen glasses using minerals as standards gave a 1s precision of +6–9% for Li, Be and B determinations where these constituents are present in significant amounts (0.49–3.33 wt% oxide); that is, the precision for determining kornerupine B content would be 10%. However, given the reproducibility reported by Grew et al. (2006) for Li, Be and B present as trace constituents, precision for all the other determinations is most likely no better than 30%. Cell parameters were refined for kornerupine in sample TK2002122104 using Ni-filtered Cu-Ka1 radiation generated at 30 kV and 15 mA on a Rigaku RAD-1B X-ray diffractometer at Okayama University. The cell parameters were refined for all the peaks by a least-squares method using the computer program ‘CellCalc’ (Miura 2003). Laser Raman spectroscopy was utilized in identifying Al2SiO5 phases at the National Institute of Polar Research (NIPR).
Description of the kornerupine-bearing lenses and associated basic rocks Kornerupine was found in two lenses several metres from one another in the hinge of the kilometre-scale synform (Fm generation, Ikeda & Kawakami 2004) exposed on Sansyoku terrace (Fig. 1). One lens is irregular in form, roughly 50 –60 cm across and surrounded by hornblende gneiss (Fig. 2a –e; samples AKR2002A, AKR2002B & TK2002122104), whereas the other is a tabular body up to 5 cm thick and 25 cm long between amphibolite and hornblende gneiss (Fig. 2f; sample I-010). Both lenses are coarser grained than surrounding hornblende gneiss (Fig. 2d–f). Lenses of pyroxenite and amphibolite are found in the hornblende gneiss with the kornerupine-bearing lenses (Fig. 2a–c).
356
T. KAWAKAMI ET AL.
Fig. 2. Photographs showing the mode of occurrence of the kornerupine-bearing rock. (a, b) General view to west shows lenses of pyroxenite, amphibolite and kornerupine–plagioclase– corundum (ruby) rock in hornblende gneiss. The relationship of (a) and (b) will be understood from the same amphibolite lens in both pictures. (c) Enlargement of pyroxenite and amphibolite lenses shown in (a). (d) Enlargement of kornerupine –plagioclase– corundum lens shown in (b). (e) Enlargement of kornerupine crystal shown in (d). (f) Tabular kornerupine-bearing lens.
Kornerupine-bearing lenses (samples I-010, AKR2002 and TK2002122104) Kornerupine. Kornerupine forms euhedral prisms up to 4 cm in diameter and 10 cm long. SIMS analyses of sample AKR2002A and AKR2002B give B less than
0.5 a.p.f.u. of 21.5 O (Table 2); that is, the mineral is kornerupine sensu stricto. Overall, the prisms in these two samples are relatively homogeneous in terms of most of the major elements (e.g. XMg ¼ 0.79– 0.82), but show minor zonation and variation from grain to grain in Li, Be and B (Fig. 3, Table 2).
¨ TZOW-HOLM COMPLEX KORNERUPINE FROM THE LU
Unit cell parameters of kornerupine determined from powder X-ray diffraction (XRD) data (Table 3) on another sample from the irregular lens (TK2002122304) are a ¼ 16.048 (5), ˚ , V ¼ 1483.9 (5) b ¼ 13.736 (2), c ¼ 6.732 (1) A 3 ˚ A . SIMS boron content and unit cell parameters of the Akarui Point kornerupine are consistent with the relationships reported by previous investigators (Werding & Schreyer 1978; Waters & Moore 1985; Grew et al. 1990; Braga et al. 2003). Using a and c cell parameters determined from a specified set of 11 lines in the powder XRD pattern (Table 3), measured FeO (wt%) and measured Al2O3 (wt%), Cooper & Hawthorne (2008) deduced three equations to calculate B and Fe3þ atoms with a precision of +0.05 atom per 21.5 O. Application of these equations to an average of the analyses listed in Table 2 gave 0.387 B and 0.118 Fe3þ per 21.5 O, which correspond to 1.80 wt% B2O3 (compared with average SIMS B2O3 of 1.98 + 0.11 wt%) and 1.25 wt% Fe2O3; that is Fe3þ constitutes 17% of total Fe. Matrix minerals in the kornerupine-bearing lenses. Major constituents of the matrix to kornerupine are euhedral ruby corundum (Cr2O3 ¼ 0.22–0.34 wt%) in tabular crystals up to 3 cm in diameter, biotite in flakes 1–5 mm across, and plagioclase. There is no textural evidence for reaction between corundum and kornerupine (Figs 2e and 9). Ti varies inversely with XMg over a limited range in biotite (XMg ¼ 0.79 –0.82, Ti ¼ 0.14–0.18 per 22 O anhydrous; Fig. 4, Table 4). Matrix plagioclase in the irregular lens ranges from bytownite (An70 –84) in AKR2002A and AKR2002B (e.g. Table 5) to labradorite (An66 –80) in TK2002122104, whereas matrix plagioclase in the tabular lens (I-010) is more sodic (An47– 53). Sapphirine, ideally (Mg7Al9)O4[Si3Al9O36], green spinel and unidentified sulphide are trace constituents in sample AKR2002. Sapphirine grains are 0.06–0.12 mm long, whereas spinel grains are finer (up to 0.05 mm) and rounded in shape; both minerals are found in matrix as well as included in plagioclase. A larger grain of spinel (up to 0.3 mm) is found adjacent to a ruby porphyroblast that is found near the kornerupine grain. Sapphirine and spinel are locally included together in a single plagioclase grain. Sapphirine composition deviates slightly from the ideal formula as a result of minor substitution of Fe2þ for Mg, Fe3þ for Al, and subtraction of the Tschermak component MgSiAl22 (Table 6). Spinel is of the hercynite –spinel series (XMg ¼ 0.49– 0.54, Cr/ (Cr þ Al þ Fe3þ) ¼ 0.04–0.08; Table 7). Overall, distribution of light elements between kornerupine and the other matrix minerals is typical of kornerupine-bearing and sapphirine-bearing
357
rocks reviewed by Grew et al. (1990, 2003, 2006). Sapphirine B content is roughly 10% of kornerupine B content (per 14 cations), a distribution characteristic of kornerupine sensu stricto. Both kornerupine and sapphirine are sinks for Be, with a slight preference for kornerupine, whereas Be contents do not exceed 1 ppm in biotite and plagioclase. Li is strongly fractionated into kornerupine relative to both biotite and sapphirine; that is, average ln KD [where KD ¼ (Li/Fe)Bt, Spr/(Li/Fe)Krn] is 21.3 to 22.3. Plagioclase and biotite contain 10–20 ppm B, ranges typical of these minerals in general (Grew 1996; the 432 ppm value found in one biotite grain is probably due to an impurity). Minerals included in kornerupine. In sample TK2002122104, kornerupine encloses tourmaline, ruby corundum, biotite, plagioclase and monazite. Tourmaline forms green, rounded grains up to 400 mm in diameter (Figs 5a and 6); it encloses andesine (An44) and is partially rimmed by andesine of similar composition (Fig. 5b). Tourmaline is calcic dravite (XMg ¼ 0.81–0.85, XCa ¼ 0.16–0.37, Ti ¼ 0.04 –0.08 per 24.5 O anhydrous without B; e.g. Table 8). Ruby is euhedral and rarely encloses anhedral crystals of biotite, kornerupine and tourmaline (Fig. 5c). Biotite inclusions are euhedral except where associated with ruby or plagioclase. The range of biotite XMg and Ti contents overlaps that of matrix biotite (Fig. 4; Table 4). There are several textures involving plagioclase; one example is thin selvedges of andesine (An34– 49) rimming ruby (Fig. 5d) and biotite, another is andesine (An46) inclusions in ruby enclosed in kornerupine. Plagioclase included in the rim of kornerupine prisms is labradorite–bytownite (An62 –82), significantly more calcic than andesine included in the core (An34 –49) and almost as calcic as plagioclase in the matrix. Zoned oligoclase (An11– 28) fills cracks in kornerupine and has no reaction relation with it. Kornerupine in sample AKR2002 includes plagioclase at the rim and the composition is An52–53, more sodic than the matrix plagioclase (An70 –84). Biotite is also found as inclusions in kornerupine. In sample I-010, matrix plagioclase is as calcic as a plagioclase included in kornerupine (around An50). Minerals replacing kornerupine. Kornerupine is partly replaced along cracks by tourmaline grains with irregular outlines (XMg ¼ 0.77 – 0.86, XCa ¼ 0.01 – 0.08, Table 7), sapphire, Ti-poor biotite (Fig. 4; Table 4), chlorite (XMg ¼ 0.80 – 0.81) and magnesite (XMg ¼ 0.78 – 0.81) (Fig. 7a and b). The secondary replacement of kornerupine is observed in both types of lenses. Rare
358
Table 2. Representative analyses of kornerupine 2002B 53 Krn-1
2002B 54 Krn-2
2002B 55 Krn-3
2002B 56 Krn-4
2002B 57 Krn-5
2002B 58 Krn-6
2002B 59 Krn-7
2002A 18 Krn-1
2002A 19 Krn-2a
2002A 21 Krn-2b
2002A 22 Krn-2c
1010 11 Krn
SiO2 TiO2 Al2O3 Cr2O3 V2O3 B2O3 BeO FeO MnO MgO CaO Na2O K2O Li2O F Cl H2O calc. Total O5 5F, Cl Total
29.44 0.11 44.56 0.08 0.05 2.19 0.01 6.60 0.26 15.81 0.05 0.06 0.00 0.13 0.00 0.02 1.20 100.6 0.00 100.6
30.08 0.11 44.14 0.14 0.03 2.05 0.00 6.69 0.26 15.87 0.05 0.10 0.00 0.08 0.05 0.01 1.18 100.8 0.02 100.8
29.85 0.10 44.23 0.10 0.03 2.10 0.00 6.73 0.17 15.75 0.07 0.08 0.01 0.09 0.00 0.00 1.21 100.5 0.00 100.5
29.54 0.15 44.16 0.19 0.00 1.92 0.00 6.65 0.19 15.58 0.06 0.07 0.01 0.08 0.00 0.00 1.20 99.8 0.00 99.8
29.49 0.11 44.70 0.10 0.02 1.87 0.00 6.64 0.23 15.89 0.04 0.08 0.00 0.08 0.02 0.01 1.19 100.5 0.01 100.5
29.35 0.11 44.03 0.11 0.06 1.95 0.00 6.53 0.23 15.74 0.07 0.08 0.00 0.08 0.01 0.00 1.19 99.5 0.00 99.5
29.02 0.16 43.60 0.06 0.06 1.76 0.00 6.77 0.24 15.75 0.06 0.07 0.00 0.08 0.00 0.02 1.18 98.8 0.00 98.8
29.78 0.09 43.84 0.06 0.02 2.02 0.01 6.54 0.28 15.84 0.07 0.05 0.00 0.11 0.00 0.01 1.20 99.9 0.00 99.9
29.81 0.07 43.73 0.18 0.02 1.97 0.02 6.90 0.29 15.82 0.06 0.03 0.00 0.11 0.00 0.00 1.20 100.2 0.00 100.2
29.33 0.02 44.18 0.13 0.04 1.97 0.01 6.87 0.22 16.18 0.05 0.02 0.00 0.08 0.00 0.00 1.20 100.3 0.00 100.3
29.72 0.07 43.61 0.20 0.03 1.97 0.02 6.65 0.27 15.78 0.06 0.05 0.01 0.11 0.06 0.00 1.17 99.8 0.02 99.8
29.29 0.08 43.75 0.08 0.04 n.a. n.a. 7.08 0.38 15.31 0.04 0.05 0.00 n.a. 0.01 0.00 1.15 97.3 0.00 97.3
T. KAWAKAMI ET AL.
Sample: Analysis number: Mineral:
21.5
21.5
21.5
21.5
21.5
21.5
21.5
21.5
21.5
21.5
21.5
21.5
Si Ti Al Cr V B Be Fe2þ Mn Mg Ca Na K Li F Cl OH calc. Total cations
3.65 0.01 6.51 0.01 0.00 0.47 0.00 0.68 0.03 2.92 0.01 0.01 0.00 0.06 0.00 0.00 1.00 14.38
3.72 0.01 6.44 0.01 0.00 0.44 0.00 0.69 0.03 2.93 0.01 0.02 0.00 0.04 0.02 0.00 0.98 14.35
3.71 0.01 6.47 0.01 0.00 0.45 0.00 0.70 0.02 2.91 0.01 0.02 0.00 0.04 0.00 0.00 1.00 14.35
3.70 0.01 6.51 0.02 0.00 0.42 0.00 0.70 0.02 2.90 0.01 0.02 0.00 0.04 0.00 0.00 1.00 14.34
3.67 0.01 6.55 0.01 0.00 0.40 0.00 0.69 0.02 2.94 0.01 0.02 0.00 0.04 0.01 0.00 0.99 14.37
3.68 0.01 6.51 0.01 0.01 0.42 0.00 0.68 0.02 2.94 0.01 0.02 0.00 0.04 0.00 0.00 1.00 14.36
3.68 0.01 6.51 0.01 0.01 0.38 0.00 0.72 0.03 2.97 0.01 0.02 0.00 0.04 0.00 0.00 1.00 14.38
3.72 0.01 6.45 0.01 0.00 0.44 0.00 0.68 0.03 2.95 0.01 0.01 0.00 0.06 0.00 0.00 1.00 14.36
3.72 0.01 6.43 0.02 0.00 0.42 0.00 0.72 0.03 2.94 0.01 0.01 0.00 0.06 0.00 0.00 1.00 14.37
3.66 0.00 6.49 0.01 0.00 0.42 0.00 0.72 0.02 3.01 0.01 0.01 0.00 0.04 0.00 0.00 1.00 14.40
3.72 0.01 6.43 0.02 0.00 0.43 0.00 0.70 0.03 2.94 0.01 0.01 0.00 0.06 0.02 0.00 0.98 14.37
3.81 0.01 6.70 0.01 0.00 n.a. n.a. 0.77 0.04 2.97 0.01 0.01 0.00 n.a. 0.00 0.00 1.00 14.33
0.81
0.80
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.81
0.79
Mg/(Fe þ Mg) ppm Li B Be
302 6812 30
182 6365 16
207 6527 14
181 5977 15
All Fe as FeO. WDS analyses at AIST and SIMS analyses at UNM. n.a., not analysed.
191 5822 15
186 6044 13
181 5466 13
260 6283 43
266 6105 58
197 6108 45
256 6129 59
n.a. n.a. n.a.
¨ TZOW-HOLM COMPLEX KORNERUPINE FROM THE LU
Number of oxygens
359
360
T. KAWAKAMI ET AL.
Fig. 3. Photomicrographs (plane-polarized light; PPL) of sections AKR2002A and AKR2002B showing the points analysed by SIMS and WDS. Li2O, BeO and B2O3 contents are plotted as functions of distance.
andalusite (identification confirmed by laser Raman spectroscopy) accompanies tourmaline in sample TK2002122104 (Fig. 7c and d). Secondary tourmaline differs from the included tourmaline in containing less Ti and more Na; that is, Na/ (Na þ Ca þ K) approaches unity, but occupancy of the X-site is lower (Fig. 6). Secondary biotite also differs from matrix, presumably primary, biotite in containing less Ti (Fig. 4). For this reason, a biotite found with andalusite and secondary tourmaline but containing 0.16 Ti per 22 O anhydrous is considered to be primary because its Ti content approaches that of matrix biotite.
Hornblende gneiss (sample I-009) This rock type mainly consists of hornblende (XMg ¼ 0.77), clinopyroxene (XMg ¼ 0.84 – 0.86; Table 9), anorthite (An97 – 98) and green spinel corrected for Fe3þ). (XMg ¼ 0.65 – 0.69; XCr ¼ Cr/(Cr þ Al þ Fe3þ) of the spinel is 0 – 0.01. Subordinate amounts of calcite, pyrite,
magnetite and haematite are also present. Clinopyroxene is partly rimmed by hornblende.
Pyroxenite (sample I-011) This rock consists of coarse-grained, black orthopyroxene segregations and domains dominated by green clinopyroxene (XMg ¼ 0.86 – 0.87) with subordinate hornblende (XMg ¼ 0.80 – 0.82) and andesine (An40 – 41). Pyrrhotite þ pentlandite are included in clinopyroxene. The clinopyroxene-dominated domain is partly altered to a dolomite-bearing assemblage consisting of hornblende, dolomite, quartz, and magnetite. Magnetite is partly replaced by a hematite, whereas ilmenite is mostly replaced by a haematite – rutile intergrowth.
Estimate of pressure – temperature conditions of metamorphism Pressure–temperature conditions of metamorphism have been estimated from a Sil–Bt–Grt leucogneiss
¨ TZOW-HOLM COMPLEX KORNERUPINE FROM THE LU
Table 3. Power X-ray diffraction kornerupine in sample TK2002122104
data
for
h
k
I
˚) dcalc (A
˚) dmeas (A
Imeas
1 2 0 2 4 0 0 0 5 3 1 5 0 5 0 0 6 1 5 5 0 5 0 0 6 5 5 5
1 0 2 2 0 4 0 2 1 1 5 3 4 1 6 2 4 5 5 5 4 5 8 0 6 7 3 9
0 0 0 1 0 0* 2* 2* 1 2 0 0* 2* 2 0 3 0* 2 0* 1 3* 2 0 4* 2* 2* 4 0
10.44 8.02 6.87 4.12 4.01 3.434 3.366 3.023 2.835 2.790 2.708 2.628 2.404 2.290 2.289 2.133 2.110 2.110 2.087 1.993 1.878 1.774 1.717 1.683 1.545 1.499 1.417 1.378
10.47 8.07 6.90 4.13 4.02 3.437 3.371 3.025 2.836 2.791 2.710 2.629 2.405 2.289 2.289 2.134 2.111 2.111 2.087 1.993 1.878 1.774 1.717 1.683 1.545 1.499 1.417 1.378
11 4 10 5 7 17 37 82 8 6 10 100 11 15 15 12 36 36 20 11 12 15 7 36 13 45 18 8
*Line for cell refinement used in calculating B and Fe3þ with equations of Cooper & Hawthorne (2008). These lines gave ˚, ˚, ˚, a ¼ 16.052(9) A b ¼ 13.735(5) A c ¼ 6.732(1) A and ˚ 3. V ¼ 1484(1) A
(sample TK2002122304; Fig. 1) using GASP geobarometers (Ghent 1976; Hodges & Spear 1982) and Grt–Bt geothermometers (Ferry & Spear 1978; Hodges & Spear 1982). The mineral assemblage in the leucogneiss is Grt þ Sil þ Bt þ Pl þ Kfs þ
Fig. 4. Compositions of matrix biotite, biotite included in kornerupine, and secondary biotite.
361
Qtz þ Ilm þ Rt þ Py þ Zrn. Sillimanite is commonly found enclosed in the plagioclase aggregates. Ilmenite that is in contact with rutile and pyrite is found as an inclusion in garnet. We selected for X-ray elemental mapping two garnet grains that has been used to estimate the P –T conditions of metamorphism (Fig. 8). Of particular interest is phosphorus (P). Because it is more resistant to intracrystalline diffusion than major elements such as Fe, Mg and Ca, P can better preserve growth zoning under upper amphibolite- to granulite-facies conditions (Hiroi et al. 1997). Both grains show P-poor cores gradually increasing in P outward (Fig. 8a and d). The cores are separated by a sharp boundary from a P-poor rim, consistent with the suggestion that intracrystalline diffusion of P is slow and P effectively preserves growth zoning. Ca preserves the growth zoning in the larger grain (Fig. 8b), but not in the smaller grain (Fig. 8e). The sharp boundary between rims and cores is not evident in the Mg maps and only very diffuse zoning is evident in the core of the larger grain (Fig. 8c and f). The rim of the smaller garnet encloses biotite grains surrounded by haloes of Mg depletion; Mg is also depleted in garnet next to matrix biotite (Fig. 8f). Both features resulted from later re-equilibrium. The P and Ca maps support our conclusion that the larger garnet grain has largely preserved growth zoning, although some intracrystalline diffusion probably had occurred. The larger garnet core also includes biotite grains, but without Mg-depletion haloes. Application of Grt – Bt geothermometers (Ferry & Spear 1978; Hodges & Spear 1982) to the biotite included in the larger garnet core and adjacent garnet gives about 570 – 630 8C (at 6.0 – 11.0 kbar). The inclusion biotite (core) and adjacent garnet (where not affected by the Mg-depletion halo) pair included in the smaller garnet rim gives 590 – 710 8C (at 6.0 – 11.0 kbar). The highest temperature estimate using Grt – Bt geothermometers and GASP geobarometers is obtained from matrix biotite, matrix plagioclase and garnet rim (740 – 875 8C, 6.0 – 9.5 kbar). The increase of temperature from the garnet core to the rim is consistent with the observation by Hiroi et al. (1983b) that garnet in the Sil – Bt – Grt leucogneiss from this locality includes staurolite and kyanite relics remaining from the prograde metamorphic stage. Sillimanite inclusions and the green spinel inclusions inferred to be the product of staurolite breakdown at the rim of the smaller garnet (Fig. 8f) are also consistent with the garnet rim growth after staurolite breakdown within the sillimanite stability field; that is, near the metamorphic peak.
362
Table 4. Representative analyses of biotite 2002A 17 Bt-1
2002A 27 Bt-2
2002A 30 Bt-3
2002A 51 Bt-1
I-010 20 Bt in matrix
I-010 21 Bt in matrix
122104 42 Bt in Krn
122104 44 Bt in Krn
I-010 19 Bt in Krn
122104 46 Bt in Krn (retrograde)
122104 65 Bt in Krn (retrograde)
SiO2 TiO2 Al2O3 Cr2O3 B2O3 BeO FeO MnO MgO CaO ZnO Na2O K2O Li2O F Cl Total O5 5F, Cl Total
36.83 2.26 19.89 0.33 0.00 0.00 8.78 0.05 17.83 0.01 n.a. 0.72 9.06 0.02 0.06 0.14 96.0 0.06 95.9
37.46 1.71 19.81 0.17 0.01 0.00 8.72 0.09 18.67 0.01 n.a. 0.79 9.04 0.01 0.08 0.16 96.7 0.07 96.7
37.32 1.91 19.96 0.20 0.01 0.00 8.47 0.12 18.70 0.00 n.a. 0.74 9.15 0.01 0.05 0.18 96.8 0.06 96.7
36.22 3.10 20.13 0.28 0.14 0.00 9.53 0.08 17.00 0.00 n.a. 0.58 9.02 0.02 0.07 0.15 96.3 0.06 96.2
36.16 1.31 19.54 0.29 n.a. n.a. 8.56 0.06 17.67 0.03 0.03 0.21 9.22 n.a. 0.27 0.15 93.5 0.15 93.3
36.11 1.27 19.72 0.16 n.a. n.a. 8.35 0.10 17.73 0.00 0.00 0.30 9.28 n.a. 0.13 0.14 93.3 0.09 93.2
37.59 1.06 20.19 0.19 n.a. n.a. 7.56 0.07 18.86 0.01 0.00 0.32 9.13 n.a. 0.56 0.13 95.7 0.27 95.4
37.73 1.01 20.14 0.24 n.a. n.a. 6.95 0.06 18.62 0.03 0.00 0.41 8.93 n.a. 0.51 0.13 94.7 0.24 94.5
36.71 1.24 20.04 0.19 n.a. n.a. 8.51 0.07 17.66 0.02 0.05 0.27 8.83 n.a. 0.15 0.11 93.8 0.09 93.8
38.19 0.04 20.75 0.07 n.a. n.a. 6.81 0.09 18.94 0.01 0.03 0.34 9.22 n.a. 0.24 0.12 94.8 0.13 94.7
37.18 0.07 20.56 0.05 n.a. n.a. 8.85 0.02 19.23 0.04 0.04 0.20 8.46 n.a. 0.43 0.08 95.2 0.20 95.0
T. KAWAKAMI ET AL.
Sample: Analysis number: Mineral:
22
22
22
22
22
22
22
22
22
22
22
Si Ti Al Cr B Be Fe2þ Mn Mg Ca Zn Na K Li F Cl Total cations
5.29 0.24 3.37 0.04 0.00 0.00 1.06 0.01 3.82 0.00 n.a. 0.20 1.66 0.01 0.03 0.03 15.69
5.34 0.18 3.32 0.02 0.00 0.00 1.04 0.01 3.96 0.00 n.a. 0.22 1.64 0.01 0.04 0.04 15.74
5.31 0.20 3.35 0.02 0.00 0.00 1.01 0.01 3.96 0.00 n.a. 0.20 1.66 0.01 0.02 0.04 15.74
5.20 0.34 3.41 0.03 0.03 0.00 1.15 0.01 3.64 0.00 n.a. 0.16 1.65 0.01 0.03 0.04 15.63
5.34 0.15 3.40 0.03 n.a. n.a. 1.06 0.01 3.89 0.01 0.00 0.06 1.74 n.a. 0.12 0.04 15.69
5.34 0.14 3.43 0.02 n.a. n.a. 1.03 0.01 3.90 0.00 0.00 0.08 1.75 n.a. 0.06 0.04 15.71
5.39 0.11 3.41 0.02 n.a. n.a. 0.91 0.01 4.03 0.00 0.00 0.09 1.67 n.a. 0.25 0.03 15.65
5.44 0.11 3.42 0.03 n.a. n.a. 0.84 0.01 4.00 0.00 0.00 0.11 1.64 n.a. 0.23 0.03 15.60
5.37 0.14 3.45 0.02 n.a. n.a. 1.04 0.01 3.85 0.00 0.01 0.08 1.65 n.a. 0.07 0.03 15.62
5.48 0.00 3.51 0.01 n.a. n.a. 0.82 0.01 4.05 0.00 0.00 0.09 1.69 n.a. 0.11 0.03 15.65
5.36 0.01 3.49 0.01 n.a. n.a. 1.07 0.00 4.13 0.01 0.00 0.06 1.56 n.a. 0.20 0.02 15.69
0.78
0.79
0.80
0.76
0.79
0.79
0.82
0.83
0.79
0.83
0.79
n.a. n.a. n.a.
n.a. n.a. n.a.
n.a. n.a. n.a.
n.a. n.a. n.a.
n.a. n.a. n.a.
n.a. n.a. n.a.
n.a. n.a. n.a.
Mg/(Fe þ Mg) ppm Li B Be
35 11 0.2
30 19 0.6
29 17 1.0
41 432 0.3
All Fe as FeO. n.a., not analysed. WDS analyses at AIST for sample 2002A; at OUS for samples 122104 and I-010. SIMS analyses at UNM.
¨ TZOW-HOLM COMPLEX KORNERUPINE FROM THE LU
Number of oxygens
363
364
T. KAWAKAMI ET AL.
Table 5. Representative analyses of plagioclase Sample: Analysis number: Mineral:
2002A 2002A 2002A 2002B 25 26 29 52 Pl-1 Pl-2 Pl-3 Pl-4
SiO2 TiO2 A12O3 Cr2O3 B2O3 BeO FeO MnO MgO CaO Na2O K2O Li2O Total
49.11 0.00 32.42 0.07 0.01 0.00 0.00 0.01 0.00 15.75 2.51 0.03 0.00 99.9
47.94 0.02 32.54 0.03 0.01 0.00 0.04 0.03 0.00 16.57 2.00 0.02 0.00 99.2
48.68 0.00 32.41 0.05 0.01 0.00 0.05 0.03 0.00 16.00 2.36 0.03 0.00 99.6
50.31 0.00 31.14 0.00 0.01 0.00 0.07 0.02 0.00 14.06 3.34 0.04 0.00 99.0
Number of oxygens
8
8
8
8
Si Ti Al Cr B Be Fe2þ Mn Mg Ca Na K Li Total cations
2.25 0.00 1.75 0.00 0.00 0.00 0.00 0.00 0.00 0.77 0.22 0.00 0.00 4.99
2.21 0.00 1.77 0.00 0.00 0.00 0.00 0.00 0.00 0.82 0.18 0.00 0.00 4.99
2.24 0.00 1.75 0.00 0.00 0.00 0.00 0.00 0.00 0.79 0.21 0.00 0.00 4.99
2.31 0.00 1.69 0.00 0.00 0.00 0.00 0.00 0.00 0.69 0.30 0.00 0.00 4.99
An ppm Li B Be
77
82
79
70
0.04 16 0.2
0.05 19 0.6
0.07 20 0.6
0.06 19 0.1
All Fe as FeO. WDS analyses at AIST and SIMS analyses at UNM.
The third garnet grain, which is not mapped, has an inclusion of biotite surrounded by plagioclase at the garnet rim. Quartz is included next to plagioclase and biotite, whereas sillimanite is found in another part of the rim of the same garnet grain; that is, garnet, biotite and quartz presumably coexisted with sillimanite, and the rim Grt –Bt– Pl–Sil –Qtz assemblage can be used for geothermobarometry. Application of geothermobarometry to this biotite, plagioclase and adjacent garnet gives P–T conditions of 770 –790 8C, 7.7–9.8 kbar. This result is reliable because biotite is not in direct contact with garnet and thus post-peak re-equilibrium was not significant.
Based on the considerations above, the rim garnet and matrix biotite pair and the rim set of the third garnet grain are interpreted to give the most reliable estimate of near-peak conditions. Nonetheless, the estimated temperatures are minima because of the possibility of some retrograde re-equilibrium between garnet and biotite. Hiroi et al. (1991) estimated 9 kbar for the minimum pressure condition at peak temperature for Akarui Point on the basis of symplectitic intergrowth of Opx þ An þ Spl around garnet in sapphirine-bearing troctolitic rocks, an estimate consistent with that obtained from matrix assemblage and inclusions in the rim of the third garnet grain.
Discussion P – T evolution of kornerupine-bearing rock The relatively high Mg/Fe ratios of the ferromagnesian silicates allow us to apply experimental data in the MgO –Al2O3 –SiO2 –H2O–B2O3 (MASHB) system as a first approximation to propose a P–T evolution of the Akarui Point kornerupine-bearing rock in conjunction with the above P– T estimates. The stability field of boron-free kornerupine was studied experimentally by Seifert (1975) and Wegge & Schreyer (1994) in the MASH system, and that of prismatine was preliminarily determined in the MASHB system by Krosse (1995) and discussed by Werding & Schreyer (1996). The experiments of Krosse (1995) roughly constrained the low-temperature stability limit of prismatine to 700–800 8C at 5–10 kbar (Werding & Schreyer 1996). Comparison of their results with the lowtemperature stability limit of boron-free kornerupine (Wegge & Schreyer 1994) implies that addition of boron does not significantly expand the stability field of kornerupine to lower temperatures between 5 and 10 kbar. Only the stability field of prismatine with 0.9 a.p.f.u. B (Krosse 1995; Werding & Schreyer 1996) is shown in Figure 9. P –T conditions estimated from the ‘inclusion-set’ lie in the stability field of boron-free kornerupine, whereas conditions estimated from the ‘matrix set’ imply decompression to 710– 720 8C and 4.1–6.2 kbar on the retrograde path, where kornerupine is no longer stable.
Kornerupine-forming and secondary replacement reactions The presence of biotite, plagioclase and rounded tourmaline inclusions in kornerupine implies that these three minerals are precursors to kornerupine.
¨ TZOW-HOLM COMPLEX KORNERUPINE FROM THE LU
365
Table 6. Representative analyses of sapphirine Sample: Analysis number: Mineral:
2002A 20 Spr-1
2002A 24 Spr-2
2002A 28 Spr-3
2002B 49 Spr-1
SiO2 TiO2 A12O3 Cr2O3 V2O3 B2O3 Fe2O3 (calc.) BeO FeO (calc.) FeO (meas.) MnO MgO CaO NiO ZnO Na2O K2O Li2O Total Number of oxygens Si Ti Al Cr V B Fe3þ Be Fe2þ Mn Mg Ca Ni Zn Na K Li Total
12.00 0.04 63.42 0.26 0.02 0.29 2.39 0.00 4.87 7.02 0.18 16.39 0.06 0.07 0.00 0.01 0.00 0.01 100.01 20 1.42 0.00 8.86 0.02 0.00 0.06 0.21 0.00 0.48 0.02 2.89 0.01 0.01 0.00 0.00 0.00 0.01 14.00
12.18 0.02 62.97 1.27 0.01 0.19 1.99 0.01 5.06 6.85 0.20 16.48 0.01 0.01 0.02 0.01 0.00 0.01 100.43 20 1.44 0.00 8.78 0.12 0.00 0.04 0.18 0.00 0.50 0.02 2.91 0.00 0.00 0.00 0.00 0.00 0.00 14.00
11.99 0.00 63.95 0.66 0.03 n.a. 1.64 0.00 4.98 6.45 0.17 16.41 0.03 0.10 0.00 0.01 0.00 n.a. 99.97 20 1.42 0.00 8.94 0.06 0.00 n.a. 0.15 0.00 0.49 0.02 2.90 0.00 0.01 0.00 0.00 0.00 n.a. 14.00
11.82 0.04 63.81 0.38 0.04 0.23 1.31 0.00 5.43 6.61 0.14 15.93 0.01 0.07 0.00 0.01 0.00 0.02 99.24 20 1.41 0.00 8.97 0.04 0.00 0.05 0.12 0.00 0.54 0.01 2.83 0.00 0.01 0.00 0.00 0.00 0.01 14.00
0.86
0.85
0.85
0.84
Mg/(Fe2þ þ Mg) ppm Li B Be
34 889 0.1*
24 588 38
n.a. n.a. 76
54 704 7
n.a., not analysed. WDS analyses at AIST and SMS analyses at UNM. FeO and Fe2O3 were calculated by stoichiometry assuming 14 cations and 20 oxygens. Total does not include FeO (meas.). *Unexpected low value could be due to ion beam missing the sapphirine grain.
Although not found as inclusions, sapphirine and spinel may have also been present before kornerupine formed, and thus could be relics of the precursor assemblage. Corundum (ruby) locally includes kornerupine and vice versa, consistent with simultaneous formation. We propose that plagioclase with moderate anorthite content
(around An34) was present before kornerupine formation, and anorthite content increased as kornerupine formed, a scenario suggested by the increase in anorthite content of plagioclase inclusions from core to rim of kornerupine prisms. A continuous kornerupine-forming reaction consistent with the above observations and using the
366
T. KAWAKAMI ET AL.
Table 7. Representative analyses of Spinel
plagioclase compositions of samples taken from the irregular lens is
consumed before tourmaline by the above reaction in the Al-poor sample, and tourmaline before sapphirine and spinel in the Al-rich sample. By analogy with the experiments of Werding & Schreyer (1978) in the MASHB system, Grew et al. (1990) inferred that kornerupine boron contents are a measure of boron activities in coexisting fluids. Applying the distribution coefficient (B2O3/H2O)Krn/(B2O3/H2O)fluid ¼ 18.7 (Werding & Schreyer 1978) to the present case, the B2O3/ H2O ratio of the fluid, if present, is calculated to be about 0.03. This is equivalent to 10.4 wt% B2O3 in the fluid. Another possible interpretation is that kornerupine formed from tourmaline under constant P–T conditions during infiltration of fluid that had reacted with mafic rocks. Decreasing Na from core to rim in tourmaline inclusions and decreasing albite component from plagioclase included in the core of kornerupine crystals to plagioclase included in the rims suggest a decrease of Na activity (or an increase in Ca activity) in the fluid as reaction between the fluid and mafic rocks progressed. The final result was the Na-poor, Ca-rich assemblage kornerupine þ An-rich plagioclase. Boron activity probably decreased as well, and eventually tourmaline was no longer stable. Introduction or remobilization of Na, K, Ca by H2O–CO2 fluids and subsequent reaction of kornerupine with these fluids explain the development of secondary biotite, chlorite (possibly after biotite), magnesite and dravite in cracks cutting kornerupine. Because rims of the kornerupine prisms are not replaced by the secondary assemblage, minerals included in kornerupine must have participated in the reaction. Plagioclase played an important role because it could have provided Ca and Na for secondary tourmaline. However, the difference in Na/Ca ratio between plagioclase inclusions in kornerupine and secondary tourmaline (Table 8) implies that there was another source of Na; that is, the interiors of the kornerupine prisms were open to Na and Ca. Introduction of Si by the fluids may be required for the andalusite formation. A possible reaction for the secondary assemblage with sapphire is
Tur þ sodic Pl ( .An34) þ Spr + Spl ! Krn þ Crn(ruby) þ calcic Pl ( , An82) + (fluid or melt): (1)
Krn þ Pl þ (Na, K in H2 O – CO2 fluid) ! Tur þ Crn (sapphire) þ Bt + Mgs þ (Ca in H2 O – CO2 fluid) (2)
Sample: Analysis number: Mineral:
2002A 42 Spl in Pl
2002A 44 Spl in Pl
0.01 0.03 56.29 6.55 0.05 3.78 19.61 23.02 0.53 12.92 0.08 0.18 0.87 0.08 0.01 101.00
0.01 0.01 56.09 7.59 0.10 1.89 21.59 23.29 0.58 11.55 0.10 0.11 0.93 0.09 0.01 100.66
Number of oxygens
4
4
Si Ti Al Cr V F3þ F2þ Mn Mg Ca Ni Zn Na K Total
0.00 0.00 1.78 0.14 0.00 0.08 0.44 0.01 0.52 0.00 0.00 0.02 0.00 0.00 3.00
0.00 0.00 1.79 0.16 0.00 0.04 0.49 0.01 0.47 0.00 0.00 0.02 0.00 0.00 3.60
Cr/(Cr þ Al þ Fe3þ) Mg/(Fe2þ þ Mg)
0.07 0.54
0.08 0.49
SiO2 TiO2 Al2O3 Cr2O3 V2O3 Fe2O3 (calc.) FeO (calc.) FeO (meas.) MnO MgO CaO NiO ZnO Na2O K2O Total
WDS analyses at AIST. FeO and Fe2O3 were calculated by stoichiometry assuming four cations and three oxygens. Total does not include FeO (meas.).
The presence of sapphirine and spinel in a sample lacking relict tourmaline (AKR2002) and their absence in a sample containing relict tourmaline (TK2004122104) could be due to variations in bulk Al/(Fe þ Mg) ratio from one sample to another. Sapphirine and spinel were probably
and that for the andalusite-bearing replacement is Krn þ Pl þ ðNa; Si in H2 O – CO2 fluidÞ ! Tur þ And þ ðCa in H2 O – CO2 fluidÞ: ð3Þ
¨ TZOW-HOLM COMPLEX KORNERUPINE FROM THE LU
367
Fig. 5. Photomicrographs of inclusion minerals and reaction textures observed in kornerupine from TK2004122104. (a) Tourmaline relic included in kornerupine PPL (b) Back-scattered electron image of the left-hand side of (a). Tourmaline further includes plagioclase (An44) and a plagioclase selvedge is observed around tourmaline. Plagioclase directly included in kornerupine at the top right has a composition of An44. (c) Tourmaline relic included in a ruby inclusion in kornerupine (PPL). (d) Plagioclase selvedge surrounding ruby inclusion in kornerupine (PPL).
The rare presence of andalusite constrains the P–T conditions not to exceed 4 kbar and 500 8C to 3 kbar and 575 8C, well outside the stability fields of B-free kornerupine and prismatine (Werding & Schreyer 1996; Fig. 9). It is not surprising that fluid infiltration triggered localized breakdown of kornerupine, which otherwise could have persisted metastably under these conditions.
Source of boron in a mafic – ultramafic environment Akarui Point is the sixth known locality worldwide where kornerupine is associated with mafic and ultramafic rocks such as anorthosite, peridotite or kimberlite (Table 1). These occurrences differ from one another, as do the scenarios proposed for formation of kornerupine and origin of boron.
Peck & Valley (1996) cited O and H isotope ratios to propose that the kornerupine-bearing Mg –Al-rich rocks at Fiskenæsset, which is the type locality for kornerupine, resulted from hydrothermal alteration of a mafic or ultramafic component of the anorthosite or of the country rock by hot seawater prior to granulite-facies metamorphism. The isotope data did not support the proposal by Herd (1973) for potassium metasomatism during later amphibolite-facies metamorphism by aqueous fluids, which could have removed boron from the metasedimentary rocks found along the contact of anorthosite with its country rock (Grew et al. 1987). Peck & Valley (1996) did not consider the source of boron. Kornerupine-bearing rocks at Moon Mountain (Farrar & Babcock 1993; Farrar 1995) are also associated with anorthosite, but unlike the Fiskenæsset complex, this anorthosite contains accessory tourmaline. Kornerupine is
Table 8. Representative analyses of tourmaline 368
Simple: Analysis number: Mineral/occurrence:
122104 69 Tur/retro
122104 74 Tur/retro
122104 87 Tur/retro
1-010 14 Tur/retro
122104 79 Tur/pro
122104 82 Tur/pro(core)
122104 85 Tur/pro
122104 90 Tur/pro
SiO2 TiO2 B2O3(calc.) Al2O3 Cr2O3 FeO MnO MgO CaO Na2O K2O Total
38.36 0.02 10.83 31.97 0.07 3.26 0.03 9.81 0.15 2.47 0.09 97.1
38.40 0.00 10.93 33.77 0.00 2.72 0.00 9.30 0.11 2.22 0.00 97.4
38.00 0.19 10.87 33.99 0.00 3.96 0.01 8.22 0.38 2.36 0.02 98.0
38.34 0.35 10.76 30.86 0.19 4.14 0.01 9.60 0.44 2.55 0.02 97.2
38.53 0.02 10.86 34.01 0.19 2.75 0.01 8.50 0.17 1.15 0.03 96.2
36.92 0.44 10.75 33.34 0.24 3.43 0.00 8.83 1.92 1.79 0.04 97.7
37.95 0.49 10.79 32.51 0.15 3.26 0.00 9.16 0.91 2.42 0.03 97.6
36.89 0.53 10.83 33.28 0.46 3.61 0.04 9.11 2.10 1.76 0.06 98.7
36.80 0.64 10.79 33.01 1.16 3.53 0.01 8.78 1.72 2.09 0.09 98.6
Number of oxygens
24.5
24.5
24.5
24.5
24.5
24.5
24.5
24.5
24.5
T site Si A1(IV) Z site A1(Z) Y site A1(Y) Cr Ti Fe Mg Mn Y-site total X site Ca Na K Vacancy Mg/(Fe þ Mg)
6.15 0.00
6.11 0.00
6.08 0.00
6.19 0.00
6.17 0.00
5.97 0.03
6.11 0.00
5.92 0.08
5.93 0.07
6.00
6.00
6.00
5.87
6.00
6.00
6.00
6.00
6.00
0.05 0.01 0.00 0.44 2.35 0.00 2.85
0.33 0.00 0.00 0.36 2.20 0.00 2.89
0.41 0.00 0.02 0.53 1.96 0.00 2.92
0.00 0.02 0.04 0.56 2.31 0.00 2.94
0.41 0.02 0.00 0.37 2.03 0.00 2.83
0.33 0.03 0.05 0.46 2.13 0.00 3.00
0.17 0.02 0.06 0.44 2.20 0.00 2.89
0.21 0.06 0.06 0.48 2.18 0.01 3.00
0.19 0.15 0.08 0.48 2.11 0.00 3.00
0.03 0.77 0.02 0.19 0.84
0.02 0.68 0.00 0.30 0.86
0.06 0.73 0.00 0.20 0.79
0.08 0.80 0.00 0.12 0.81
0.03 0.36 0.01 0.61 0.85
0.33 0.56 0.01 0.10 0.82
0.16 0.76 0.01 0.08 0.83
0.36 0.55 0.01 0.08 0.82
0.30 0.65 0.02 0.03 0.82
All Fe as FeO. Boron calculated by stoichiometry (3 boron a.p.f.u.). WDS analyses at OUS. Retro, retrograde: pro, prograde. Formulae of tourmaline normalized on the basis of 15 cations exclusive of B, Na, Ca and K.
T. KAWAKAMI ET AL.
122104 52 Tur/retro
¨ TZOW-HOLM COMPLEX KORNERUPINE FROM THE LU
Fig. 6. Compositional plots of precursor and secondary tourmalines. Precursor tourmaline is found included in kornerupine (e.g. Fig. 5a–c), whereas secondary tourmaline is found as a breakdown product (e.g. Fig. 7). (a) (Na þ Ca þ K) per formula unit v. Na/(Na þ Ca þ K). (b) Mg/(Mg þ Fe) v. X-site vacancy per formula unit.
found in the rim of a peridotite body, probably a cumulate, at the margin of the anorthosite sill. Farrar & Miller (2005) suggested that the anorthositic magma and peridotite cumulates were contaminated by boron-rich metasediments through which the magma had passed. Kornerupine was found in a single boulder in the Kittila¨ wilderness, which was presumed to have originated in a narrow belt of hornblende gneiss, amphibolite and hornblendite metamorphosed under upper amphibolite- to lower granulite-facies conditions. The presence of chromite indicates a mafic–ultramafic igneous precursor (Haapala et al. 1971), whereas sapphirine and corundum are clearly metamorphic (Schumacher 1984). At Lherz, kornerupine was thought by some investigators (e.g. Monchoux 1972; Vielzeuf & Kornprobst 1984) to have resulted from contact metasomatism at high temperature by peridotitic magma on country rock, but the mechanism of
369
emplacement of the peridotite has been debated (e.g. Hall & Bennett 1979; Choukroune 1980; Minnigh et al. 1980), which leaves some ambiguity in the interpretation of how kornerupine formation is related to peridotite at Lherz. Golberg et al. (1986) reported tourmaline in marble in the country rock at Lherz, a finding suggesting a possible source of boron for kornerupine. In contrast to the other occurrences, kornerupine in the alkremite xenolith enclosed in kimberlite at Moses Rock is a late phase associated with secondary chlorite and formed from breakdown of garnet and spinel in the presence of aqueous fluid (Padovani & Tracy 1981), which could have originated in crustal rocks (Smith 1995). In summary, the source of boron at four of the five localities (no information is available on Kittila¨) could be sedimentary rocks or fluids that had extracted boron from them. The paragenesis at Akarui Point has more in common with those at Fiskenæsset, Moon Mountain and Kittila¨ than with those at Lherz and Moses Rock, but application of the models proposed for these three localities to Akarui Point is not straightforward because of differences between these four occurrences. In contrast to the anorthosite intrusive rocks at Fiskenæsset and Moon Mountain, the mafic rocks at Akarui Point and Kittila¨ are not part of a well-defined igneous complex, but instead are highly deformed and metamorphosed layers whose igneous origin is inferred from bulk composition and from the presence of chromite. Chrome enrichment of ultramafic rocks on Akarui Point is indicated by preliminary study of ultramafic lenses exposed to the north of the localities where kornerupine is found (sample TK2002122205; T. Kawakami, unpubl. data). The Akarui Point and Kittila¨ kornerupine and corundum contain significant Cr, as does kornerupine at some localities in the Fiskenæsset complex (e.g. Petersen et al. 1980), but not the type locality (Herd 1973; Grew et al. 1987); the presence of Cr in kornerupine and associated minerals at Moon Mountain was not mentioned. Herd (1973) attributed the localized Cr enrichments in sapphirine and kornerupine associated with the Fiskenæsset complex to residual Cr-rich fluids associated with the magma, but this explanation may not be appropriate to Akarui Point because we find no evidence of a well-defined intrusive rock. Another critical difference is that kornerupine at Akarui Point probably had a tourmaline precursor, whereas there is no evidence for a tourmaline precursor at the three other localities, where tourmaline is secondary. A tourmaline precursor implies that boron was introduced early in the metamorphic evolution prior to formation of kornerupine and not during a later metamorphic event as Grew et al. (1987) suggested for Fiskenæsset
370
T. KAWAKAMI ET AL.
Fig. 7. Photomicrographs showing the mode of occurrence of secondary tourmaline from TK2004122104. (a) Biotite, sapphire, magnesite and secondary tourmaline found in the altered part of kornerupine. Plagioclase developed in the crack has a sodic composition, An11-28. It should be noted that no reaction relationship is observed between sodic plagioclase and kornerupine (PPL). (b) Same as (a), but in cross-polarized light. (c) Secondary biotite, secondary tourmaline, secondary andalusite and plagioclase (An38) found in the altered part of kornerupine (PPL). (d) Same as (c), but in cross-polarized light.
kornerupine. This conclusion would also be valid even if kornerupine formed from tourmaline during infiltration of fluid that increased Ca/Na ratio, an alternative interpretation considered above for Akarui Point. Three scenarios are consistent with information currently available for Akarui Point. One is that aqueous fluids containing boron released by prograde metamorphism of sedimentary rocks, such as garnet – biotite gneiss (Fig. 1), could have infiltrated precursors of the Akarui Point kornerupinebearing rocks. Because muscovite incorporates significant boron even in non-pegmatitic environments (6–270 ppm; Grew 1996), consumption of muscovite through prograde metamorphic reactions could give a boron-bearing fluid even if tourmaline remained stable (e.g. Kawakami & Ikeda 2003). Introduction of boron-bearing fluid into the Mgand Al-rich rocks could have been facilitated by pressure gradients during deformation. These processes would have taken place under conditions
where tourmaline is stable (i.e. at temperatures below those of the upper amphibolite facies (e.g. Kawakami 2001)), and would have formed precursor tourmaline. Subsequent temperature increase resulted in the breakdown of tourmaline via reaction (1) to form kornerupine. The boron-bearing fluid could have also mobilized Cr, but from ultramafic and mafic intrusive rocks rather than sediments. Several investigators have invoked hydrothermal mobilization of Cr from ultramafic rocks to explain occurrences of chromian tourmaline, but Torres-Ruiz et al. (2003) presented evidence for immobile Cr and metasomatic introduction of boron to explain the origin of chromian tourmaline in metacarbonate rocks the Nevado-Fila´bride complex, Spain. In the second scenario, kornerupine and tourmaline formed sequentially under nearly constant P– T conditions during the infiltration of fluid whose composition evolved through reaction with mafic rocks; that is, tourmaline breakdown
¨ TZOW-HOLM COMPLEX KORNERUPINE FROM THE LU
371
Table 9. Representative analyses of constituent minerals of I-009 and I-011 Sample: Analysis number: Mineral:
I-009 2-03 Pl
I-009 2-04 Hb
I-009 2-13 Spl
I-009 2-20 Cpx
I-011 07 Pl
I-011 2-24 Act
I-011 3-03 Cpx
SiO2 TiO2 Al2O3 Cr2O3 Fe2O3(calc.) FeO(calc.) FeO(meas.) MnO MgO CaO Na2O K2O Total
43.67 n.d. 36.45 n.d.
41.00 0.32 17.23 n.d.
50.98 n.d. 5.89 n.d.
57.56 n.d. 26.93 n.d.
56.21 n.d. 2.49 n.d.
54.06 n.d. 1.47 n.d.
n.d. n.d. n.d. 20.09 0.16 n.d. 100.37
7.89 n.d. 14.66 12.86 1.79 1.47 97.22
n.d. n.d. 62.22 0.29 5.62 14.88 19.94 0.42 17.06 n.d. 0.14 n.d. 100.63
4.90 n.d. 14.63 24.82 n.d. n.d. 101.22
n.a. n.d. n.a. 8.30 6.64 0.14 99.57
7.49 n.d. 19.27 12.81 n.d. n.d. 98.27
4.45 n.d. 15.97 22.92 0.50 n.d. 99.37
4
3
8
0.93 n.d. 0.13 n.d.
2.59 n.d. 1.43 n.d.
7.81 n.d. 0.41 n.d.
0.99 n.d. 0.03 n.d.
0.07 n.d. 0.40 0.48 n.d. n.d.
n.a. n.d. n.d. 0.40 0.58 0.01
0.87 n.d. 3.99 1.91 n.d. n.d.
0.07 n.d. 0.44 0.45 0.02 n.d.
2.01
5.00
14.98
2.00
0.82
0.86
Number of oxygens
8
Si Ti Al Cr Fe3þ Fe2þ Mn Mg Ca Na K
2.01 n.d. 1.98 n.d.
5.98 0.04 2.96 n.d.
n.d. n.d. n.d. 0.99 0.01 n.d.
0.96 n.d. 3.18 2.01 0.51 0.27
n.d. n.d. 1.89 0.01 0.11 0.32 0.01 0.66 n.d. 0.01 n.d.
5.00
15.90
3.00
Total An Cr/(Cr þ Al þ Fe3þ) Mg/(Fe2þ þ Mg)
23
99
23
3
41 0.77
0.003 0.67
0.84
All Fe as FeO except for Spl. EDS analyses at ANU. n.d, not detected; n.a., not analysed. FeO and Fe2O3 in Spl were calculated by stoichiometry assuming four cations and three oxygens. Total for Spl does not include FeO (meas.).
was not controlled by temperature increase as in the first scenario. Instead, the appearance of kornerupine in place of tourmaline is controlled by the evolution of fluid composition, which was B-bearing initially, but then became depleted in B and Na (or gained Ca). The third scenario is based on the model of Peck & Valley (1996) for the Fiskenæsset anorthosite complex: localized hydrothermal alteration of mafic– ultramafic rocks by heated seawater prior to burial and metamorphism. This scenario is consistent with these ultramafic rocks being fractured blocks of an ophiolite complex that Hiroi et al. (1991) postulated as the missing oceanic crust between older continents, which are now represented by the Yamato –Belgica and Rayner complexes. Mobilization of Cr by brines is plausible given the role of these fluids in emerald formation (e.g. Franz & Morteani 2002); brines
could have also mobilized Be that was subsequently incorporated in kornerupine. Distinguishing between the three alternative scenarios would require stable isotope data. Peck & Valley (1996) cited low d18O and high dD as evidence of exchange with heated seawater; boron isotopes might constrain possible source rocks for boron (e.g. Leeman & Sisson 1996). However, isotope studies are beyond the scope of the present paper, so we present the three scenarios as equally viable explanations for the presence of Cr- and B-bearing lenses intercalated with mafic rocks. We are grateful to the crews of icebreaker Shirase and members of JARE-33, -42 and -44, especially to M. Ishikawa, Y. Kawano and T. Kawasaki for their support in the field. Y. Osanai and K. Shiraishi are thanked for their support in preparing for the field
372
T. KAWAKAMI ET AL.
Fig. 8. X-ray elemental mapping of two garnet grains from a Sil–Bt –Grt leucogneiss (TK2002122304) A sharp boundary (s.b.) is preserved between the core and the rim in the two P maps; a more diffuse boundary is evident in one Ca map. (a) P map of larger garnet. (b) Ca map of larger garnet. (c) Mg map of larger garnet. (d) P map of smaller garnet. (e) Ca map of smaller garnet. (f) Mg map of smaller garnet. survey. T. Hokada is thanked for assistance in laser Raman spectroscopy at NIPR, T. Tsujimori for WDS analyses at OUS, and M. Cooper for providing in advance of publication his method for calculation of B and Fe3þ contents
in kornerupine from powder X-ray diffraction data. We are also grateful to A. Liebscher and an anonymous reviewer for their reviews of an earlier version of the manuscript, P. Tuisku for constructive comments, and
Fig. 9. (a) The result of P– T estimates and estimated P– T evolution of kornerupine-bearing rock. The bold dashed line represents experimentally determined P– T location of the reaction Grt + Hb ! Opx þ An þ Spl + Hb (Hiroi et al. 1991). HS, Grt–Bt geothermometer and GASP geobarometer of Hodges & Spear (1982); FS, Grt– Bt geothermometer of Ferry & Spear (1978), Ghent, GASP geobarometer of Ghent (1976). (b) A P– T diagram showing the stability field of prismatine with 0.9 a.p.f.u. B (Krosse 1995; Werding & Schreyer 1996), tourmaline breakdown reactions and reactions involving muscovite. The bold dashed line is the same reaction line as given in (a). Tur-out (Ryoke) is empirically derived high-temperature stability limit of tourmaline in Ryoke metapelitic rocks (Kawakami 2004). Breakdown reaction line of alkali-free dravite (þboron-rich fluid) is from Werding & Schreyer (1984), and that of dravite is from Schreyer & Werding (1997). Dehydration and melting reactions involving muscovite are from Johannes & Holtz (1996). Some reactions have been omitted to simplify the diagram. Akfs, alkali-feldspar.
¨ TZOW-HOLM COMPLEX KORNERUPINE FROM THE LU M. Satish-Kumar for editorial assistance. This study was partly done when T.K. was at Okayama University and AIST. Field survey in Antarctica was financially supported by the Grant-in-Aid for JSPS Fellows (No. 5864) to T.K. E.S.G.’s research, including the SIMS analyses, was supported by US National Science Foundation grant OPP-0228842 to the University of Maine.
References B RAGA , R., C ALLEGARI , A., M ESSIGA , B., O TTOLINI , L., R ENNA , M. R. & T RIBUZIO , R. 2003. Origin of prismatine from the Sondalo granulites (Central Alps, northern Italy). European Journal of Mineralogy, 15, 393– 400. C ARSON , C. J., H AND , M. & D IRKS , P. H. G. M. 1995. Stable coexistence of grandidierite and kornerupine during medium pressure granulite facies metamorphism. Mineralogical Magazine, 59, 327 –339. C HOUKROUNE , P. 1980. Comment on ‘Quenching; an additional model for emplacement of the lherzolite at Lers (French Pyrenees)’. Geology, 8, 514. C OOPER , M. A. 1997. The crystal chemistry of kornerupine. MSc thesis, University of Manitoba, Winnipeg. C OOPER , M. A. & H AWTHORNE , F. C. 2008. The crystal chemistry of the kornerupine–prismatine series. IV. Complete chemical formulae from electronmicroprobe analysis and X-ray powder diffraction. Canadian Mineralogist (in press). D YAR , M. D., W IEDENBECK , M., R OBERTSON , D. ET AL . 2001. Reference minerals for the microanalysis of light elements. Geostandards Newsletter, 25, 441–463. E VENSEN , J. M., L ONDON , D. & W ENDLANDT , R. F. 1999. Solubility and stability of beryl in granitic melts. American Mineralogist, 84, 733–745. F ARRAR , S. S. 1995. Mg– Al– B-rich facies associated with the Moon Mountain metanorthosite sill, southeastern Adirondacks, NY. Geological Society of America, Abstracts with Programs, 27, 42– 43. F ARRAR , S. S. & B ABCOCK , L. G. 1993. A sapphirine þ kornerupine-bearing hornblende spinel peridotite associated with Adirondack anorthositic sill. Geological Society of America, Abstracts with Programs, 25, A-265. F ARRAR , S. S. & M ILLER , J. M. 2005. Southeastern Adirondack anorthositic sills and dikes with included ultramafic lenses. Geological Society of America, Abstracts with Programs, 37, 22. F ERRY , J. M. & S PEAR , F. S. 1978. Experimental calibration of the partitioning of Fe and Mg between biotite and garnet. Contributions to Mineralogy and Petrology, 66, 113– 117. F RANZ , G. & M ORTEANI , G. 2002. Be-minerals: Synthesis, stability, and occurrence in metamorphic rocks. In: G REW , E. S. (ed.) Beryllium: Mineralogy, Petrology and Geochemistry. Mineralogical Society of America, Reviews in Mineralogy and Geochemistry, 50, 551–589. F RASER , G., M C D OUGALL , I., E LLIS , D. J. & W ILLIAMS , I. S. 2000. Timing and rate of isothermal decompression in Pan-African granulites from
373
Rundva˚gshetta, East Antarctica. Journal of Metamorphic Geology, 18, 441–454. G HENT , E. D. 1976. Plagioclase–garnet– Al2SiO5 – quartz: a potential geobarometer –geothermometer. American Mineralogist, 61, 710– 714. G REW , E. S. 1988. Kornerupine at the Sar-e-Sang, Afghanistan, whiteschist locality: Implications for tourmaline–kornerupine distribution in metamorphic rocks. American Mineralogist, 73, 345–357. G REW , E. S. 1996. Borosilicates (exclusive of tourmaline) and boron in rock-forming minerals in metamorphic environments. In: A NOVITZ , L. M. & G REW , E. S. (eds) Boron: Mineralogy, Petrology and Geochemistry. Mineralogical Society of America, Reviews in Mineralogy, 33, 387–502. G REW , E. S., H ERD , R. K. & M ARQUEZ , N. 1987. Boronbearing kornerupine from Fiskenæsset, West Greenland: a re-examination of specimens from the type locality. Mineralogical Magazine, 51, 695– 708. G REW , E. S., C HERNOSKY , G., W ERDING , G., A BRAHAM , K., M ARQUEZ , N. & H INTHORNE , J. R. 1990. Chemistry of kornerupine and associated minerals, a wet chemical, ion microprobe, and X-ray study emphasizing Li, Be, B and F contents. Journal of Petrology, 31, 1025– 1070. G REW , E. S., R AO , A. T., R AJU , K. K. V. S. ET AL . 2003. Prismatine and ferroho¨gbomite-2N2S in granulitefacies Fe-oxide lenses in the Eastern Ghats Belt at Venugopalapuram, Vizianagaram district, Andhra Pradesh, India: do such lenses have a tourmalineenriched lateritic precursor? Mineralogical Magazine, 67, 1081–1098. G REW , E. S., Y ATES , M. G., S HEARER , C. K., H AGERTY , J. J., S HERATON , J. W. & S ANDIFORD , M. 2006. Beryllium and other trace elements in paragneisses and anatectic veins of the ultrahigh-temperature Napier Complex, Enderby Land, East Antarctica: The role of sapphirine. Journal of Petrology, 47, 859– 882. H AAPALA , I., S IIVOLA , J., O JANPERA¨ , P. & Y LETYINEN , V. 1971. Red corundum, sapphirine and kornerupine from Kittila¨, Finnish Lapland. Bulletin of the Geological Society of Finland, 43, 221 –231. H ALL , C. A. & B ENNETT , V. C. 1979. Significance of lherzolite at the Etang de Lherz, central Pyrenees, southern France. Earth and Planetary Science Letters, 45, 349–354. H ERD , R. K. 1973. Sapphirine and kornerupine occurrences within the Fiskenæsset complex. Grønlands Geologiske Undersøgelse Rapport, 51, 65–71. H ERD , R. K., W INDLEY , B. F. & G HISLER , M. 1969. The mode of occurrence and petrogenesis of the sapphirine-bearing and associated rocks of West Greenland. Grønlands Geologiske Undersøgelse Rapport, 24, 1 –44. H IROI , Y., S HIRAISHI , K., N AKAI , Y., K ANO , T. & Y OSHIKURA , S. 1983a. Geology and petrology of Prince Olav Coast, East Antarctica. In: O LIVER , R. L., J AMES , P. R. & J AGO , J. B. (eds) Antarctic Earth Science. Australian Academy of Science, Canberra, A.C.T., 32–35. H IROI , Y., S HIRAISHI , K., Y ANAI , K. & K IZAKI , K. 1983b. Aluminum silicates in the Prince Olav and Soya Coasts, East Antarctica. Memoirs of National
374
T. KAWAKAMI ET AL.
Institute for Polar Research, Special Issue, 28, 115– 131. H IROI , Y., S HIRAISHI , K., M OTOYOSHI , Y., K ANISAWA , S., Y ANAI , K. & K IZAKI , K. 1986. Mode of occurrence, bulk chemical compositions, and mineral textures of ultramafic rocks in the Lu¨tzowHolm Complex, East Antarctica. Memoirs of National Institute for Polar Research, Special Issue, 43, 62– 84. H IROI , Y., S HIRAISHI , K., M OTOYOSHI , Y. & K ATSUSHIMA , T. 1987. Progressive metamorphism of calc-silicate rocks from the Prince Olav and Soya Coasts, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 1, 73– 97. H IROI , Y., S HIRAISHI , K. & M OTOYOSHI , Y. 1991. Late Proterozoic paired metamorphic complexes in East Antarctica, with special reference to the tectonic significance of ultramafic rocks. In: T HOMSON , M. R. A., C RAME , J. A. & T HOMSON , J. W. (eds) Geological Evolution of Antarctica, Cambridge University Press, Cambridge, 83– 87. H IROI , Y., M OTOYOSHI , Y., E LLIS , D. J., S HIRAISHI , K. & K ONDO , Y. 1997. The significance of phosphorus zonation in garnet from high grade pelitic rocks: A new indicator of partial melting. In: R ICCI , C. A. (ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antartica, Siena, 73–77. H ODGES , K. V. & S PEAR , F. S. 1982. Geothermometry, geobarometry and the Al2SiO5 triple point at Mt. Moosilauke, New Hampshire. American Mineralogist, 67, 1118–1134. I KEDA , T. & K AWAKAMI , T. 2004. Structural analysis of the Lu¨tzow-Holm Complex in Akarui Point, East Antarctica, and overview of the complex. Polar Geoscience, 17, 22– 34. J OHANNES , W. & H OLTZ , W. 1996. Petrogenesis and Experimental Petrology of Granitic Rocks. Springer, Berlin. K AWAKAMI , T. 2001. Tourmaline breakdown in the migmatite zone of the Ryoke metamorphic belt, SW Japan. Journal of Metamorphic Geology, 19, 61–75. K AWAKAMI , T. 2004. Tourmaline and boron as indicators of the presence, segregation and extraction of melt in pelitic migmatites: examples from the Ryoke metamorphic belt, SW Japan. Transactions of the Royal Society of Edinburgh: Earth Sciences, 95, 111– 124. K AWAKAMI , T. & I KEDA , T. 2003. Depletion of whole-rock boron controlled by the breakdown of tourmaline and retrograde formation of borosilicates in the Yanai area, Ryoke metamorphic belt, SW Japan. Contributions to Mineralogy and Petrology, 145, 131–150. K AWASAKI , T., I SHIKAWA , M. & M OTOYOSHI , Y. 1993. A preliminary report on cordierite-bearing assemblages from Rundva˚gshetta region, Lu¨tzow-Holm Bay, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 6, 47– 56. K ELLY , N. M. & H ARLEY , S. L. 2004. Orthopyroxene – corundum in Mg–Al-rich granulites from the Øygarden Islands, East Antarctica. Journal of Petrology, 45, 1481– 1512. K RETZ , R. 1983. Symbols for rock forming minerals. American Mineralogist, 68, 277–279.
K ROSSE , S. 1995. Hochdrucksynthese, Stabilita¨t und Eigenschaften der Borsilikate Dravit und Kornerupin sowie Darstellung und Stabilita¨tsverhalten eines neuen Mg–Al-Borates. Doctor’s thesis, Ruhr-Universita¨t, Bochum. L EEMAN , W. P. & S ISSON , V. B. 1996. Geochemistry of boron and its implications for crustal and mantle processes. In: A NOVITZ , L. M. & G REW , E. S. (eds) Boron: Mineralogy, Petrology and Geochemistry. Mineralogical Society of America, Reviews in Mineralogy, 33, 645– 708. M INNIGH , L. D., VAN C ALSTEREN , P. W. C. & DEN T EX , E. 1980. Quenching; an additional model for emplacement of the lherzolite at Lers (French Pyrenees). Geology, 8, 18–21. M IURA , H. 2003. CellCalc: A unit cell parameter refinement program on Windows computer. Journal of the Crystallographic Society of Japan, 45, 145– 147 [in Japanese]. M ONCHOUX , P. 1972. Roches a` sapphirine au contact des lherzolites pyre´ne´ennes. Contributions to Mineralogy and Petrology, 37, 47– 64. M OTOYOSHI , Y. 1986. Prograde and progressive metamorphism of the granulite facies Lu¨tzow-Holm Bay region, East Antarctica. DSc thesis, Hokkaido University Sapporo, Hokkaido, Japan. M OTOYOSHI , Y. & I SHIKAWA , M. 1997. Metamorphic and structural evolution of granulites from Rundva˚gshetta, Lu¨tzow-Holm Bay, East Antarctica. In: R ICCI , C. A. (ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antartica, Siena, 65–72. O SANAI , Y., T OYOSHIMA , T., O WADA , M. ET AL . 2004. Antarctic Geological Map Series, Sheet 39 Skallen (revised edition). National Institute of Polar Research, Tokyo. P ADOVANI , E. R. & T RACY , R. J. 1981. A pyrope–spinel (alkremite) xenolith from Moses Rock Dike: first known North American occurrence. American Mineralogist, 66, 741–745. P ECK , W. H. & V ALLEY , J. W. 1996. The Fiskenæsset anorthosite complex; stable isotope evidence for shallow emplacement into Archean ocean crust. Geology, 24, 523– 526. P ETERSEN , O. V., J OHNSEN , O. & J ENSEN , A. 1980. Giant crystals of kornerupine. Mineralogical Record, 11, 93–96 P OST , N. J., H ENSEN , B. J. & K INNY , P. D. 1997. Two metamorphic episodes during a 1340– 1180 Ma convergent tectonic event in the Windmill Islands, East Antarctica. In: R ICCI , C. A. (ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antartica, Siena, 157– 161. S CHREYER , W. & W ERDING , G. 1997. High-pressure behaviour of selected boron minerals and the question of boron distribution between fluids and rocks. Lithos, 41, 251–266. S CHUMACHER , J. C. 1984. Corundum– sapphirine –spinel amphibolites containing relict Zn– Al chromite: possible metamorphosed mafic igneous rocks, Kittila¨, Finnish Lapland. Fortschritte der Mineralogie, Beiheft, 62, 228–229. S EIFERT , F. 1975. Stability of sapphirine: A study of the aluminous part of the system MgO–Al2O3 –SiO2 – H2O. Journal of Geology, 82, 173– 204.
¨ TZOW-HOLM COMPLEX KORNERUPINE FROM THE LU S HIRAISHI , K., H IROI , Y. & O NUKI , H. 1984. Orthopyroxene-bearing rocks from the Tenmondai and Naga-iwa Rocks in the Prince Olav Coast, East Antarctica: First appearance of orthopyroxene in progressive metamorphic sequence. Memoirs of National Institute for Polar Research, Special Issue, 33, 126–144. S HIRAISHI , K., H IROI , Y., E LLIS , D. J., F ANNING , C. M., M OTOYOSHI , Y. & N AKAI , Y. 1992. The first report of a Cambrian orogenic belt in East Antarctica—An ion microprobe study of the Lu¨tzow-Holm Complex. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Science. Terra, Tokyo, 67– 73. S HIRAISHI , K., E LLIS , D. J., H IROI , Y., F ANNING , C. M., M OTOYOSHI , Y. & N AKAI , Y. 1994. Cambrian orogenic belt in East Antarctica and Sri Lanka: Implications for Gondwana assembly. Journal of Geology, 102, 47–65. S MITH , D. 1995. Chlorite-rich ultramafic reaction zones in Colorado Plateau xenoliths: recorders of sub-Moho hydration. Contributions to Mineralogy and Petrology, 121, 185–200. T ORRES -R UIZ , J., P ESQUERA , A. & L O´ PEZ S A´ NCHEZ V IZCAI´ NO , V. 2003. Chromian tourmaline and associated Cr-bearing minerals from the Nevado-Fila´bride complex (Betic Cordilleras, SE Spain). Mineralogical Magazine, 67, 517–533. V IELZEUF , D. & K ORNPROBST , J. 1984. Crustal splitting and the emplacement of Pyrenean lherzolites and
375
granulites. Earth and Planetary Science Letters, 67, 87–96. W ATERS , D. J. & M OORE , J. M. 1985. Kornerupine in Mg–Al-rich gneisses from Namaqualand, South Africa: mineralogy and evidence for late-metamorphic fluid activity. Contributions to Mineralogy and Petrology, 91, 369– 382. W EGGE , S. & S CHREYER , W. 1994. Boron-free kornerupine: its upper pressure stability limit in the system (MASH). European MgO– Al2O3 – SiO2 –H2O Journal of Mineralogy, 6, 67–75. W ERDING , G. & S CHREYER , W. 1978. Synthesis and crystal chemistry of kornerupine in the system MgO– Al2O3 – SiO2 –B2O3 – H2O. Contributions to Mineralogy and Petrology, 67, 247– 259. W ERDING , G. & S CHREYER , W. 1984. Alkali-free tourmaline in the system MgO–Al2O3 –B2O3 –SiO2 – H2O. Geochimica et Cosmochimica Acta, 48, 1331– 1344. W ERDING , G. & S CHREYER , W. 1996. Experimental studies on borosilicates and selected borates. In: A NOVITZ , L. M. & G REW , E. S. (eds) Boron: Mineralogy, Petrology and Geochemistry. Mineralogical Society of America, Reviews in Mineralogy, 33, 117– 163. Y ANAI , K., K IZAKI , K., S HIRAISHI , K., H IROI , Y. & K ANISAWA , S. 1984. Antarctic geological map series sheet 20, Akarui Point and Naga-iwa Rock. National Institute of Polar Research, Tokyo.
Sapphirine 1 quartz association in garnet: implication for ultrahigh-temperature metamorphism at Rundva˚gshetta, Lu¨tzow-Holm Complex, East Antarctica YASUTAKA YOSHIMURA1, YOICHI MOTOYOSHI2 & TOMOHARU MIYAMOTO3 1
Department of Natural Environmental Science, Kochi University, Akebono-cho 2-5-1, Kochi 780-8520, Japan
2
National Institute of Polar Research, Kaga 1-9-10, Itabashi-ku, Tokyo 173-8515, Japan (e-mail:
[email protected]) 3
Department of Earth and Planetary Science, Kyushu University, Hakozaki 6-10-1, Fukuoka 812-8581, Japan Abstract: We report the occurrence of sapphirine þ quartz association within garnet porphyroblast in the garnet– orthopyroxene–sillimanite granulite (Grt– Opx–Sill granulite) from Rundva˚gshetta in the Lu¨tzow-Holm Complex, East Antarctica. The granulites in the study area show a characteristic mineral assemblage consisting of orthopyroxene þ sillimanite þ quartz. The presence of sapphirine and quartz inclusions within garnet in the sapphirine-bearing Grt–Opx–Sill granulite suggests that metamorphic conditions changed from the stability field of orthopyroxene þ sillimanite þ quartz to that of sapphirine þ quartz during the garnet growth. Peak metamorphic temperature conditions of about 1000–1100 8C are obtained by ternary feldspar thermometry for these granulites. Similar temperatures were also estimated from the Al-in-orthopyroxene geothermometer. The granulites are also characterized by coarse-grained garnet, being partly surrounded by a fine-grained symplectite composed of orthopyroxene and cordierite, whereas sapphirine þ cordierite symplectitic intergrowth occurs in the matrix. These textures imply that the area underwent isothermal decompression subsequent to ultrahightemperature metamorphism. The sapphirine-bearing Grt–Opx– Sill granulite is likely to be the restitic product of partial melting and shows signs of segregation and movement of melt.
High-grade metamorphic rocks are dominant components of the lower crust, and their process of formation is important in understanding the tectonics and the evolution of continental crust. In recent years, extreme crustal metamorphism at ultrahigh-temperature (UHT) conditions has received much attention as a critical phenomenon in deep crustal evolution and crust–mantle interaction processes (Harley 2004, and references therein). Many recent studies have attempted to integrate petrographic information with thermodynamic modelling, especially using an internally consistent thermodynamic dataset, and have succeeded in obtaining precise phase equilibria constraints during UHT metamorphism (see Kelsey 2008, and references therein). Moreover, UHT metamorphic rocks have been reported from various metamorphic terranes, and their importance has been recognized for understanding the thermal gradient of the lower crust and upper mantle in Earth history (Brown 2006). In a Gondwana context, UHT metamorphic rocks have been reported from the Pan-African granulite terranes, such as the Trivandrum granulite block (Morimoto et al. 2004) and the Madurai block (Raith et al.
1997; Sajeev et al. 2004, 2006; Tateishi et al. 2004; Tsunogae & Santosh 2006; Santosh & Sajeev 2006) in southern India, the Highland Complex in Sri Lanka (Kriegsman & Schumacher 1999; Sajeev & Osanai 2004a, b; Osanai et al. 2006), the Andriamena Unit in north–central Madagascar (Concalves et al. 2004) and the Lu¨tzow-Holm Complex in East Antarctica (Motoyoshi & Ishikawa 1997; Kawakami & Motoyoshi 2004; Yoshimura et al. 2004). The metamorphic complex of East Antarctica, especially the Lu¨tzow-Holm Complex, has been considered as a key area for understanding the evolution of the Gondwana supercontinent, as the complex formed an integral segment of East Gondwana. This region is also important in terms of understanding the tectonothermal events associated with the Pan-African orogeny, as well as in modelling the development and evolution of continental crust. High-grade metamorphic rocks of upper amphibolite to granulite facies in the Lu¨tzow-Holm Complex are exposed along the Prince Olav Coast and at Lu¨tzow-Holm Bay, East Antarctica. The metamorphism in this complex is characterized by clockwise P –T path as indicated by the presence
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 377 –390. DOI: 10.1144/SP308.18 0305-8719/08/$15.00 # The Geological Society of London 2008.
378
Y. YOSHIMURA ET AL.
of prograde kyanite and staurolite inclusions within garnet or plagioclase (Hiroi et al. 1983; Motoyoshi 1986; Motoyoshi & Ishikawa 1997). The metamorphic event has been dated at c. 520 –550 Ma by U –Pb zircon sensitive high-resolution ion microprobe (SHRIMP) (Shiraishi et al. 1992, 1994) and at c. 500 –560 Ma by monazite electron microprobe (EMP) dating (Hokada & Motoyoshi 2006). Recent results show that the peak metamorphism of the granulite terrane in the eastern part of the Lu¨tzow-Holm Complex reached UHT conditions (Motoyoshi & Ishikawa 1997; Kawakami & Motoyoshi 2004; Yoshimura et al. 2004; Kawasaki & Motoyoshi 2006). This paper describes the sapphirine-bearing UHT metamorphic rocks and host garnet –biotite gneiss from Rundva˚gshetta. A sapphirine þ quartz association is reported here for the first time from the Lu¨tzow-Holm Complex, East Antarctica. We document the petrological, textural and mineral reaction characteristics, which provide conclusive evidence for UHT metamorphism, and discuss the P–T evolution of the region.
Geological outline The Lu¨tzow-Holm Complex is exposed along the Prince Olav Coast and at Lu¨tzow-Holm Bay, and is bounded by the Late Proterozoic Rayner Complex to the east (Fig. 1). The Lu¨tzow-Holm Complex shows a marked westward progression in metamorphic grade from upper amphibolite facies to granulite facies (e.g. Shiraishi et al. 1989a, b; Hiroi et al. 1991). Salient and well-recognized results from this region are as follows. (1) From the presence of prograde kyanite and staurolite inclusions within garnet or plagioclase, the metamorphism in this complex was inferred to be characterized by a clockwise P– T path (Hiroi et al. 1983; Motoyoshi & Ishikawa 1997). (2) The thermal maximum is presumed to be around Rundva˚gshetta in the southern part of Lu¨tzow-Holm Bay (Motoyoshi 1986). (3) Peak metamorphism is estimated to have occurred at c. 520– 550 Ma (Shiraishi et al. 1992, 1994). (4) The highgrade terrane subsequently underwent isothermal decompression as indicated by the formation of cordierite-bearing retrograde symplectite (Kawasaki et al. 1993; Motoyoshi & Ishikawa 1997; Fraser et al. 2000). (5) The western part of this complex underwent UHT metamorphism (Motoyoshi & Ishikawa 1997; Yoshimura et al. 2004; Kawasaki & Motoyoshi 2006). Rundva˚gshetta is located in the southwestern part of the Lu¨tzow-Holm Complex within the granulitefacies zone, about 70 km SW of Syowa Station (Fig. 1a). According to Motoyoshi & Ishikawa (1997), clinopyroxene–orthopyroxene–hornblende
granulite, garnet–biotite gneiss and garnet– sillimanite gneiss are the major rock types in this region (Fig. 1b). Minor amounts of cordierite bearing garnet–orthopyroxene and Grt–Opx–Sill granulite are also present as lenses and layers within the garnet–biotite and garnet–sillimanite gneiss. The cordierite-bearing garnet–orthopyroxene granulite and Grt–Opx–Sill granulite in this area were earlier investigated by Kawasaki et al. (1993), Motoyoshi & Ishikawa (1997) and experimentally by Kawasaki & Motoyoshi (2006). Field studies at Rundva˚gshetta carried out by JARE-40 (Japanese Antarctic Research Expedition, 1998–1999 season) investigated the sapphirinebearing Grt– Opx–Sill granulite in detail, which were then assigned to the cordierite-bearing garnet–orthopyroxene and Grt –Opx– Sill granulite category of Motoyoshi & Ishikawa (1997).
Field relations and petrography Sapphirine-bearing Grt – Opx – Sill granulite The sapphirine-bearing Grt–Opx–Sill granulite contains garnet, orthopyroxene and sillimanite, with subordinate sapphirine, cordierite, biotite, plagioclase, alkali feldspar and quartz. The rocks occur within garnet–biotite gneiss and garnet–sillimanite gneiss as layers of several centimetres to 10 cm thickness (Fig. 2), and are laterally continuous. The sapphirinebearing Grt–Opx–Sill granulite is generally massive, and in places exhibits well-developed gneissosity. Coarse-grained garnet, orthopyroxene and sillimanite define the porphyroblast assemblage (Fig. 3a and b). Coarse-grained garnets are partly surrounded by a fine-grained symplectite composed of orthopyroxene and cordierite (Fig. 3c and d). Sapphirine occurs as a symplectitic intergrowth with cordierite and plagioclase in the matrix (Fig. 4a), and is also found as inclusions in coarse garnet crystals (Fig. 4b and c). Coarse-grained garnets commonly have sapphirine, sillimanite, feldspar and quartz inclusions. Sapphirine inclusions are associated with quartz in garnet (Fig. 4d). Two stages of orthopyroxene formation were recognized, the coarsegrained crystals and a symplectite with cordierite. Cordierite is generally fine-grained and occurs as a symplectitic intergrowth with sapphirine and plagioclase, or orthopyroxene. Minor amounts of biotite occur in the margin of garnet and orthopyroxene within the matrix, and show symplectitic intergrowth with plagioclase or quartz (Figs 3b and 4a).
Garnet – biotite gneiss Garnet –biotite gneiss is widespread in this area and displays a well-developed gneissosity.
˚ GSHETTA UHT METAMORPHISM AT RUNDVA
379
Fig. 1. (a) Progressive metamorphism in the Lu¨tzow-Holm Complex, eastern Dronning Maud Land, East Antarctica. (b) Simplified geological map of Rundva˚gshetta, Lu¨tzow-Holm Complex (modified after Ishikawa et al. 1994; Motoyoshi & Ishikawa 1997).
380
Y. YOSHIMURA ET AL.
than that in the coarse-grained type. The rock contains well-developed leucosomes of various sizes that are concordant with, semi-concordant with, or discordant to the gneissic structure. Stromatic and schlieren migmatite structures are developed in both rock types. The fine-grained type exhibits gneissose structure. Both rock types consist predominantly of garnet, biotite, plagioclase, K-feldspar and quartz, with minor sillimanite. Porphyroblastic garnets in the fine-grained type commonly have idioblastic to subidioblastic and poikiloblastic textures with euhedral feldspar and quartz inclusions, and almost totally clear rims. Fig. 2. Mode of occurrence of sapphirine-bearing Grt– Opx–Sill granulite from Rundva˚gshetta.
Garnet– biotite gneiss can be subdivided into two main types based on grain size of garnet and modal ratio of biotite: a coarse-grained type with coarse grained garnets and a fine-grained type with relatively finer-grained garnets. The modal ratio of biotite in the fine-grained type is higher
Garnet – sillimanite gneiss Garnet –sillimanite gneiss is a common lithology in the study area. The rocks contain garnet, sillimanite, K-feldspar, quartz and graphite, with minor amount of plagioclase. Hydrous minerals are markedly absent. Garnet–sillimanite gneiss is moderately well foliated, with the garnet and sillimanite grains aligned parallel to the gneissic fabric. The
Fig. 3. Mineral assemblages and reaction textures in the sapphirine-bearing Grt–Opx–Sill granulite from Rundva˚gshetta. (a) Photograph of polished slab of sapphirine-bearing Grt–Opx–Sill granulite. (b) Bt þ Qtz symplectitic intergrowth around garnet. (c, d) backscattered electron image of the Opx þ Crd symplectite around garnet. Grt, garnet; Opx, orthopyroxene; Sil, sillimanite; Crd, cordierite; Bt, biotite; Pl, plagioclase; Kfs, K-feldspar; Qtz, quartz.
˚ GSHETTA UHT METAMORPHISM AT RUNDVA
381
Fig. 4. Photomicrographs and backscattered electron images of sapphirine-bearing Grt – Opx – Sill granulite from Rundva˚gshetta. (a) Symplectitic intergrowth of sapphirine þ cordierite þ plagioclase in the matrix. (b) Sapphirine-bearing composite inclusions witnin garnet porphyroblast. (c) Backscattered electron image of the sapphirine þ cordierite þ plagioclase symplectite. (d) Backscattered electron image of the sapphirine-bearing composite inclusions within the garnet. (Note that sapphirine and quartz inclusions are in contact in the garnet porphyroblast.) Spr, sapphirine.
core portions of garnet grains contain numerous fine-grained inclusions.
Mineral chemistry Analytical techniques Constituent minerals were analysed by EPMA, and the results for garnet, orthopyroxene, sapphirine, biotite, cordierite and plagioclase are given in Tables 1–3. The analyses were performed using a wavelength-dispersive (WDS) electron probe microanalyser (JEOL JXA-8600MX) housed at Kochi University. For quantitative analyses, an acceleration voltage of 15 kV, probe current of 1.5 1028 A, and beam diameter of 1 or 5 mm were used with oxideZAF correction applied to the X-ray intensity data. For compositional mapping, an acceleration voltage of 15 kV, probe current of 7.5 1027 A, dwell time of 50 ms and beam diameter of 5 mm were used.
Garnet Garnet core compositions from sapphirine-bearing Grt–Opx–Sill granulite, as well as the coarse- and finegrained types of garnet–biotite gneiss are in the range of Prp50 – 55Alm40 – 45Sps1 – 2Grs4 – 5, Prp38 – 43Alm52 – 59 Sps1 – 2Grs2 – 3 and Prp33 – 35Alm58 – 60Sps1 – 2Grs4 – 5, respectively (Fig. 5). Garnet in the garnet–sillimanite gneiss has core composition of Prp48 – 50Alm44 – 48 Sps1 – 2Grs3 – 4. Pyprope contents of garnet in sapphirine-bearing Grt–Opx –Sill granulite and garnet– sillimanite gneiss are higher than that of the garnet–biotite gneiss. In garnet–biotite gneiss, Prp contents of the coarse-grained type are higher than those of the fine-grained type (Fig. 5). The P2O5 content of garnets in garnet–sillimanite gneiss is the highest, reaching maximum values of about 0.2 wt%. Sapphirine-bearing Grt–Opx–Sill granulite and garnet–biotite gneiss contain a maximum of about 0.13 and 0.08 wt% P2O5, respectively (Table 1). Garnets in garnet–sillimanite and the coarse-grained type of garnet–biotite gneiss show the
382
Y. YOSHIMURA ET AL.
Table 1. Representative electron microprobe results for garnet and orthopyroxene Grt (O ¼ 12)
Mineral: Rock type: Sample:
Spr–Grt –Opx granulite
Opx (O ¼ 6) Grt –Bt gneiss
Spr– Grt – Opx granulite 99013111A-3-2
99013111A-3
99013111A-3-2
99020103-1
GAR22
GAR18
GAR43
40.83 0.01 23.22 0.03 19.55 0.16 0.00 15.27 2.02 0.00 0.00 0.06 101.13
40.80 0.03 23.11 0.01 19.14 0.18 0.00 15.54 1.80 0.00 0.00 0.08 100.69
SiO2 TiO2 Al2O3 Cr2O3 FeO* MnO NiO MgO CaO Na2O K2O P2O5 Total Si Ti Al Cr Fe Mn Ni Mg Ca Na K P Total
2.982 0.000 1.999 0.002 1.194 0.010 0.000 1.662 0.158 0.000 0.000 0.004 8.011
2.986 0.001 1.994 0.000 1.172 0.011 0.000 1.696 0.141 0.000 0.000 0.005 8.007
XMg Prp Alm Sps Grs
0.582 54.98 39.49 0.32 5.22
0.591 56.16 38.80 0.36 4.67
39.11 0.01 22.12 0.16 27.67 0.36 0.00 8.58 3.12 0.00 0.00 0.04 101.16 2.986 0.001 1.991 0.009 1.767 0.023 0.000 0.976 0.255 0.000 0.000 0.003 8.010 0.356 32.31 Ens 58.48 Fs 0.76 Wo 8.45
OPX18R
OPX33R
OPX154
50.08 0.05 9.15 0.07 14.60 0.12 0.00 25.92 0.03 0.00 0.00 0.01 100.05
50.22 0.08 9.10 0.08 14.37 0.04 0.00 26.07 0.06 0.01 0.00 0.02 100.04
50.41 0.10 8.96 0.10 15.20 0.02 0.00 25.90 0.03 0.00 0.00 0.01 100.73
1.794 0.001 0.386 0.002 0.437 0.004 0.000 1.384 0.001 0.000 0.000 0.000 4.010
1.796 0.002 0.384 0.002 0.430 0.001 0.000 1.390 0.002 0.000 0.000 0.001 4.008
1.797 0.003 0.377 0.003 0.453 0.001 0.000 1.376 0.001 0.000 0.000 0.000 4.010
0.760 75.94 24.00 0.07
0.764 76.29 23.59 0.12
0.752 75.18 24.76 0.06
* Total Fe as FeO.
highest Y2O3 contents, reaching maximum values of about 0.14 wt%.
decreasing towards rims. XMg values are lowest in cores and increase towards rims.
Orthopyroxene
Sapphirine
The Al2O3 content of orthopyroxene is high in sapphirine-bearing Grt –Opx– Sill granulite, reaching a maximum of about 9.5 wt% (Fig. 6). The XMg value of orthopyroxene is generally high: for sapphirine-bearing Grt –Opx–Sill granulite it is about 0.72 –0.78. A compositional map of orthopyroxene from sapphirine-bearing Grt –Opx– Sill granulite (Fig. 7) shows that Mg and Al decrease toward rim. Orthopyroxene is generally compositionally zoned, with Al values highest in cores and gradually
XMg of sapphirine ranges from 0.8 to 0.85 in sapphirine-bearing Grt– Opx–Sill granulite (Table 2). Sapphirine is poor in Cr. The Fe3þ content of sapphirine was obtained by stoichiometric calculation using the method of Higgins et al. (1979). Fe3þ of sapphirine in sapphirinebearing Grt–Opx –Sill granulite has a maximum value of 0.11 a.p.f.u. The XMg values of sapphirine inclusions within garnet in sapphirine-bearing Grt – Opx –Sill granulite tend to be slightly higher than
˚ GSHETTA UHT METAMORPHISM AT RUNDVA
383
Table 2. Representative electron microprobe results for sapphirine and cordierite Rock type:
Spr– Grt – Opx granulite
Mineral:
Spr
Sample:
99013111A-3
Crd 99013111A-3-2
99013111A-3
SPR9
SPR24
SPR1
SPR6
UN1
SiO2 TiO2 Al2O3 Cr2O3 FeO* MnO NiO MgO CaO Na2O K2O P2O5 Total
13.27 0.04 63.10 0.13 6.08 0.04 0.00 16.62 0.01 0.00 0.00 0.00 99.28
13.38 0.04 63.76 0.08 6.92 0.01 0.00 16.58 0.00 0.01 0.00 0.03 100.82
13.36 0.01 63.63 0.25 6.66 0.00 0.00 16.86 0.01 0.00 0.00 0.00 100.79
13.26 0.01 63.48 0.21 6.03 0.04 0.00 17.01 0.03 0.00 0.00 0.00 100.08
50.32 0.01 33.84 0.01 2.30 0.00 0.00 12.36 0.02 0.03 0.00 0.03 98.91
Si Ti Al Fe3þ Cr Fe2þ Mn Ni Mg Ca Na K P Total
1.578 0.003 8.843 0.000 0.012 0.604 0.004 0.000 2.945 0.001 0.000 0.000 0.000 13.991
1.572 0.003 8.829 0.018 0.008 0.661 0.001 0.000 2.904 0.000 0.003 0.000 0.003 14.003
1.569 0.001 8.810 0.028 0.024 0.627 0.000 0.000 2.953 0.001 0.000 0.000 0.000 14.013
1.565 0.001 8.828 0.023 0.019 0.572 0.004 0.000 2.992 0.004 0.001 0.000 0.000 14.011
5.006 0.001 3.967 – 0.001 0.191 0.000 0.000 1.833 0.002 0.005 0.000 0.003 11.008
XMg
0.830
0.815
0.825
0.839
0.905
*Total Fe as FeO.
those for symplectitic intergrowth with cordierite and plagioclase in matrix.
Biotite XMg of biotite ranges from 0.72 to 0.84 in sapphirine-bearing garnet – orthopyroxene granulite, from 0.62 to 0.68 in coarse-grained garnet – biotite gneiss, and from 0.52 to 0.6 in fine-grained garnet –biotite gneiss (Table 3). TiO2 contents are in almost the same range for all rock types, with a maximum of about 6 wt%. The XMg value of biotite inclusions within garnet in garnet – biotite gneiss tends to be slightly higher than in the matrix. In sapphirine-bearing Grt–Opx –Sill granulite, biotite contains a maximum of about 1.35 wt% F (Table 3). In the garnet –biotite gneiss, fluorine
contents of the coarse-grained and fine-grained types are 0.3–0.9 and 0.2–0.6 wt%, respectively. Biotite or phlogopite formed as a product of UHT metamorphism is characterized by high fluorine contents of up to 8 wt% (Motoyoshi 1998). The biotites in the sapphirine-bearing Grt–Opx –Sill granulite and garnet–biotite gneiss from Rundva˚gshetta, however, have lower fluorine contents compared with typical UHT metamorphic biotite (e.g. Motoyoshi & Hensen 2001).
Cordierite Cordierite is present in sapphirine-bearing Grt– Opx –Sill granulite, where it forms a symplectitic intergrowth with orthopyroxene around the coarsegrained garnet, and also forms symplectitic textures
Table 3. Representative microprobe results for biotite and feldspars
Rock type: Sample:
Si Ti Al Cr Fe2þ Mn Ni Mg Ca Na K P Zn Ba Total XMg Ca/(Ca þ Na) Ab An Or *Total Fe as FeO.
Plg
Spr – Grt– Opx granulite 99013111A-3
Grt –Bt gneiss
99013111A-3-2
Spr –Grt –Opx granulite
Kfs
MP
Grt – Bt gneiss
99020103-1
99013111A-3
99013111A-3-2
99020103-1
99013111A-3-2
BT8(þPlg)
BT1
BT7R
BT14
PLG2
PLG15
PLG8
KFS5
MP(PL)3
MP(K)4
38.12 5.99 15.84 0.04 10.35 0.00 0.00 16.61 0.00 0.05 9.65 0.01 0.02 0.00 0.86 20.36 97.18
38.23 5.59 16.27 0.07 9.64 0.06 0.00 16.61 0.00 0.07 8.68 0.00 0.00 0.00 1.30 20.55 95.97
39.10 4.75 16.32 0.11 8.91 0.02 0.00 17.81 0.01 0.08 9.63 0.01 0.00 0.00 1.19 20.50 97.43
37.00 5.47 16.33 0.26 15.72 0.00 0.00 12.51 0.00 0.07 9.91 0.01 0.04 0.00 0.58 20.25 97.65
61.90 0.03 23.81 0.08 0.10 0.01 0.00 0.00 5.10 8.56 0.28 0.22 – 0.06 – – 100.15
61.39 0.03 23.85 0.00 0.04 0.00 0.00 0.02 6.26 7.95 0.29 0.09 – 0.00 – – 99.91
59.35 0.02 25.26 0.05 0.07 0.01 0.00 0.00 7.73 6.95 0.46 0.03 – 0.05 – – 99.99
64.66 0.01 18.55 0.01 0.14 0.03 0.00 0.01 0.20 1.18 15.01 0.22 – 0.54 – – 100.56
61.36 0.04 23.99 0.02 0.01 0.00 0.00 0.00 5.97 8.10 0.17 0.03 – 0.06 – – 99.74
64.26 0.03 18.34 0.00 0.02 0.00 0.00 0.01 0.07 1.00 15.37 0.07 – 0.62 – – 99.80
5.483 0.647 2.686 0.005 1.245 0.000 0.000 3.562 0.000 0.014 1.770 0.001 0.002 0.000 15.416
5.526 0.608 2.771 0.007 1.165 0.007 0.000 3.579 0.001 0.020 1.601 0.000 0.000 0.000 15.287
5.564 0.508 2.737 0.012 1.060 0.003 0.000 3.779 0.001 0.023 1.748 0.001 0.000 0.000 15.436
5.440 0.605 2.829 0.030 1.932 0.000 0.000 2.742 0.000 0.019 1.858 0.002 0.005 0.000 15.461
0.741
0.754
0.781
0.587
2.743 0.001 1.243 0.003 0.004 0.000 0.000 0.000 0.242 0.736 0.016 0.008
2.731 0.001 1.250 0.000 0.001 0.000 0.000 0.001 0.298 0.685 0.016 0.003
2.654 0.001 1.331 0.002 0.003 0.000 0.000 0.000 0.370 0.603 0.026 0.001
2.974 0.000 1.006 0.000 0.006 0.001 0.000 0.001 0.010 0.105 0.881 0.009
2.732 0.001 1.259 0.001 0.000 0.000 0.000 0.000 0.285 0.700 0.009 0.001
2.984 0.001 1.004 0.000 0.001 0.000 0.000 0.001 0.003 0.090 0.910 0.003
0.001 4.997
0.000 4.989
0.001 4.992
0.010 5.003
0.001 4.990
0.011 5.009
0.25 74 24 2
0.30 69 30 2
0.38 60 37 3
0.09 11 1 88
0.29 70 29 1
0.04 9 0 91
Y. YOSHIMURA ET AL.
SiO2 TiO2 Al2O3 Cr2O3 FeO* MnO NiO MgO CaO Na2O K2 O P2O5 ZnO BaO F O¼ Total
Bt
384
Mineral:
˚ GSHETTA UHT METAMORPHISM AT RUNDVA
385
Fig. 5. (a) Pyrope (Prp)–almandine (Alm)– grossular (Grs) diagram of garnet compositions of rocks from Rundva˚gshetta. (b) Pyrope (Prp)–almandine (Alm)– spessartine (Sps) diagram of garnet compositions of rocks from Rundva˚gshetta.
with sapphirine and plagioclase. XMg values for cordierite in the symplectitic corona with orthopyroxene and symplectite with sapphirine and plagioclase are about 0.95 and 0.9, respectively.
Plagioclase Plagioclase in sapphirine-bearing Grt–Opx –Sill granulite has anorthite contents between 23 and 30, but lower values for the coarse-grained garnet–biotite gneiss (An16 – 20). In the fine-grained garnet –biotite gneiss, the An content of plagioclase shows wide variations of An26 – 39. Plagioclase
Fig. 6. Variations of Al2O3 and XMg of orthopyroxene in the sapphirine-bearing Grt–Opx– Sill granulite.
inclusions within porphyroblastic garnet in garnet– biotite gneiss tend to have distinctly higher An values than in the matrix, at An40 – 47. The Or content of plagioclase has almost same value, about Or0.7 – 3, for both types of garnet– biotite gneiss.
Discussion Metamorphic conditions and P– T trajectory The high-grade metamorphic rocks at Rundva˚gshetta represent various lithofacies. The anhydrous lithofacies occurring in this region, such as sapphirine-bearing Grt–Opx –Sill granulite and garnet– sillimanite gneiss, in the near-absence of any hydrous minerals, suggest high-temperature
Fig. 7. Mg and Al compositional maps of orthopyroxene in the sapphirine-bearing Grt–Opx–Sill granulite. H represents higher concentration and L lower concentration of each element.
386
Y. YOSHIMURA ET AL.
conditions of metamorphism. In sapphirine-bearing Grt–Opx –Sill granulite from Rundva˚gshetta, coarse-grained garnet, orthopyroxene, sillimanite and quartz were the stable assemblage at temperatures near peak metamorphism. The coexistence of orthopyroxene, sillimanite and quartz in this rock indicates that these rocks experienced UHT metamorphism. Another important piece of evidence for UHT metamorphism is the presence of Spr þ Qtz in the garnet porphyroblast. In particular, the sapphirine þ quartz association in the sapphirine-bearing Grt– Opx–Sill granulite indicates UHT conditions at over 1050 8C (Bertrand et al. 1991). In addition, UHT metamorphism is supported by the presence of highly aluminous orthopyroxene (Al2O3 contents up to 9.5 wt%) in the sapphirine-bearing Grt–Opx –Sill granulite. A temperature above 1050 8C is estimated by applying the Al2O3 isopleths of Fe–Mg orthopyroxene in FMAS assemblages (Harley & Motoyoshi 2000) (Fig. 8). The K-feldspar in sapphirine-bearing Grt–Opx –Sill granulite shows a mesoperthitic texture with coarse exsolution lamellae of Nafeldspar within host K-feldspar. Hokada (2001)
showed that an An–Ab–Or ternary feldspar solvus is useful in estimating the peak metamorphic temperature of UHT metamorphic rocks. The reintegrated chemical composition of the mesoperthites was obtained using the modal ratio of host lamellae and their respective chemical compositions following the techniques described by Yoshimura et al. (2000). The solvus used is that of Kroll et al. (1993), which is a modified version of that of Elkins & Grove (1990). Although the mesoperthite lamellae (Pl and Or lamellae) could have re-equilibrated compositionally at comparatively low temperatures, the compositions of many feldspars, after restoration to a single phase, plot on the solvus curve indicating a thermal maxima at about 1000–1100 8C for P ¼ 8 kbar and about 1100 8C for P ¼ 10 kbar. The results indicate that sapphirine-bearing Grt–Opx–Sill granulite in this region underwent UHT metamorphism at temperatures of at least 1000 8C, and that the peak metamorphic temperature was as high as 1100 8C (Fig. 9). We infer the following stages for the metamorphic history and P–T trajectory of Rundva˚gshetta. In the prograde stage, the metamorphic conditions changed from the stability field of kyanite to that
Fig. 8. Al2O3 isopleths of Fe– Mg orthopyroxene in FMAS assemblages (modified after Harley & Motoyoshi 2000). Shaded area is the range of Al2O3 content of orthopyroxene from Rundva˚gshetta.
˚ GSHETTA UHT METAMORPHISM AT RUNDVA
Fig. 9. An– Ab– Or ternary feldspar diagram of calculated original alkali-feldspar composition, from plagioclase and orthoclase lamella compositions in samples from Rundva˚gshetta. Ternary miscibility gaps are calculated for P ¼ 8 kbar (continuous line) and P ¼ 10 kbar (dashed line) using the method of Kroll et al. (1993).
Fig. 10. Probable decompressional P– T trajectory of the sapphirine-bearing Grt– Opx– Sill granulite from Rundva˚gshetta.
387
388
Y. YOSHIMURA ET AL.
of sillimanite because kyanite has been recognized as inclusions in porphyroblastic garnet within garnet – sillimanite gneiss at Rundva˚gshetta (Motoyoshi et al. 1985, 1989; Motoyoshi & Ishikawa 1997). At or near the peak metamorphic conditions, probably represented by orthopyroxene þ sillimanite þ quartz assemblage, a sapphirine þ quartz association also formed and was trapped within garnet, which still continued to grow. The area subsequently underwent nearisothermal decompression as indicated by various decompression textures such as cordierite-bearing symplectites surrounding garnet as well as in the matrix (Fig. 10). Sapphirine formation is recognized in two stages, namely sapphirine þ quartz in garnet and symplectitic intergrowth with cordierite and plagioclase in the matrix. We assume that there is only nominal difference in temperature conditions between the two generations of sapphirine because the XMg values of sapphirine are similar. The sapphirine þ cordierite symplectite suggests decompression and inferred reactions in
the model FMAS are as follows: orthopyroxeneþsillimanite ¼ sapphirineþcordierite garnetþquartz ¼ orthopyroxeneþcordierite: The metamorphic conditions and P – T path of the Lu¨tzow-Holm Complex estimated in previous studies (Lu¨tzow-Holm Complex as compiled by Hiroi et al. (1987) and Skallevikshalsen as deduced by Yoshimura et al. (2004)) are shown in Figure 11. The estimated peak metamorphic P – T conditions of Rundva˚gshetta clearly define UHT conditions and a clockwise P – T trajectory with steep isothermal decompression near the peak UHT metamorphism. However, contrasting P – T trajectories have been reported from the neighbouring regions of East Gondwana, which include: (1) isothermal decompression after peak metamorphism (Motoyoshi & Ishikawa 1997; Raith et al. 1997; Sajeev et al. 2004; this study); (2) isobaric cooling after peak metamorphism
Fig. 11. Summary of P– T conditions of metamorphism and P –T paths of Rundva˚gshetta and other regions in the Lu¨tzow-Holm Complex. The reaction curves are partly modified after Harley (1998) and Spear et al. (1999). Biotite melting reactions are modified from Vielzeuf & Holloway (1988) and Hensen & Osanai (1994).
˚ GSHETTA UHT METAMORPHISM AT RUNDVA
(Concalves et al. 2004; Sajeev & Osanai 2004a, b); (3) anticlockwise P – T path (Santosh & Sajeev 2006). It is still unclear whether the UHT metamorphism in all these tectonically related terrane resulted from the same heat source (magma?) followed by exhumation histories under different tectonic regimes or the UHT metamorphism itself is temporally different. It is generally considered that the high-grade metamorphism in East Gondwana was caused by the Pan-African collision between East and West Gondwana. Not only is a more detailed study on P – T evolution essential, but it is also necessary to investigate in detail the temporal and spatial relations with igneous rocks to infer the tectonic setting and heat source for UHT metamorphism. We would like to sincerely thank all members of JARE-40, and all crew members of the icebreaker Shirase. Thanks are also due to Y. Osanai, T. Kawasaki and Owada for their helpful discussions. The authors would also like to thank M. Santosh for helpful discussions and for critical reading of the manuscript. S. Dasgupta, S. Baba and K. Sajeev are thanked for constructive reviews.
References B ERTRAND , P., E LLIS , D. J. & G REEN , D. H. 1991. The stability of sapphirine–quartz and hypersthene– sillimanite– quartz assemblages: an experimental investigation in the system FeO– MgO–Al2O3 –SiO2 under H2O and CO2 conditions. Contributions to Mineralogy and Petrology, 108, 55–71. B ROWN , M. 2006. Duality of thermal regimes is the distinctive characteristic of plate tectonics since the Neoarchean. Geology, 34, 961 –964. C ONCALVES , P., N ICOLLET , C. & M ONTEL , J. 2004. Petrology and in situ U–Th– Pb monazite geochronology of ultrahigh-temperature metamorphism from the Andriamena mafic unit, north–central Madagascar. Significance of a petrographical P –T path in a polymetamorphic context. Journal of Petrology, 45, 1923–1957. E LKINS , L. T. & G ROVE , T. L. 1990. Ternary feldspar experiments and thermodynamic models. American Mineralogist, 75, 544 –559. F RASER , G., M C D OUGALL , I., E LLIS , D. J. & W ILLIAMS , I. S. 2000. Timing and rate of isothermal decompression in Pan-African granulites from Rundva˚gshetta, East Antarctica. Journal of Metamorphic Geology, 18, 441– 454. H ARLEY , S. L. 1998. On the occurrence and characterization of ultrahigh-temperature (UHT) crustal metamorphism. In: T RELOAR , P. J. & O’B RIEN , P. (eds) What Drives Metamorphism and Metamorphic Reactions? Geological Society, London, Special Publications, 138, 75– 101. H ARLEY , S. L. 2004. Extending our understanding of ultrahigh temperature crustal metamorphism. Journal of Mineralogical and Petrological Sciences, 99, 140–158.
389
H ARLEY , S. L. & M OTOYOSHI , Y. 2000. Al zoning orthopyroxene in a sapphirine quartzite: evidence for .1120 8C UHT metamorphism in the Napier Complex, Antarctica, and implications for the entropy of sapphirine. Contributions to Mineralogy and Petrology, 138, 293– 307. H ENSEN , B. J. & O SANAI , Y. 1994. Experimental study of dehydration melting of F-biotite in model pelitic compositions. Mineralogical Magazine, 58A, 410– 411. H IGGINS , J. B., R IBBE , P. H. & H ERD , R. K. 1979. Sapphirine. I. Crystal chemical contributions. Contributions to Mineralogy and Petrology, 68, 349– 356. H IROI , Y., S HIRAISHI , K., Y ANAI , K. & K IZAKI , K. 1983. Aluminum silicates in the Prince Olav and Soya Coasts, East Antarctica. Memoirs of National Institute for Polar Research, Special Issue, 28, 115–131. H IROI , Y., S HIRAISHI , K., M OTOYOSHI , Y. & K ATHUSHIMA , T. 1987. Progressive metamorphism of calc-silicate rocks from the Prince Olav and Soˆya Coasts, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 1, 73–97. H IROI , Y., S HIRAISHI , K. & M OTOYOSHI , Y. 1991. Late Proterozoic paired metamorphic complexes in East Antarctica, with special reference to the tectonic significance of ultramafic rocks. In: T HOMSON , M. R. A., C RAME , J. A. & T HOMSON , J. W. (eds) Geological Evolution of Antarctica. Cambridge University Press, Cambridge, 83– 87. H OKADA , T. 2001. Feldspar thermometry in ultrahightemperature metamorphic rocks: Evidence of crustal metamorphism attaining 1100 8C in the Archean Napier Complex, East Antarctica. American Mineralogist, 86, 932–938. H OKADA , T. & M OTOYOSHI , Y. 2006. Electron microprobe technique for U–Th– Pb and REE chemistry of monazite, and its implications for pre-, peak- and post-metamorphic events of the Lu¨tzow-Holm Complex and Napier Complex, East Antarctica. Polar Geoscience, 19, 118–151. I SHIKAWA , M., M OTOYOSHI , Y., F RASER , G. L. & K AWASAKI , T. 1994. Structural evolution of Rundva˚gshetta region, Lu¨tzow-Holm Bay, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 7, 69– 89. K AWAKAMI , T. & M OTOYOSHI , Y. 2004. Timing of attainment of the spinel þ quartz coexistence in garnet– sillimanite leucogneiss from Skallevikshalsen, Lu¨tzow-Holm complex, East Antarctica. Journal of Mineralogical and Petrological Sciences, 99, 311– 319. K AWASAKI , T. & M OTOYOSHI , Y. 2006. Experimental constraints on the decompressional P– T path of Rundva˚gshetta Granulites, Lu¨tzow-Holm complex, East Antarctica. In: F U¨ TTERER , D. K., D AMASKE , D., K LEINSCHMIDT , G., M ILLER , H. & T ESSENSOHN , F. (eds) Antarctica: Contributions to Global Earth Sciences. Springer, Berlin, 23– 36. K AWASAKI , T., I SHIKAWA , M. & M OTOYOSHI , Y. 1993. A preliminary report on cordierite-bearing assemblages from Rundva˚gshetta, Lu¨tzow-Holm Bay, East Antarctica: evidence for a decompressional P –T path? Proceedings of NIPR Symposium on Antarctic Geosciences, 6, 47– 56.
390
Y. YOSHIMURA ET AL.
K ELSEY , D. E. 2008. On ultrahigh-temperature crustal metamorphism. Gondwana Research, 13, 1 –29. K RIEGSMAN , L. & S CHUMACHER , J. C. 1999. Petrology of sapphirine-bearing and associated granulites from central Sri Lanka. Journal of Petrology, 40, 1211–1239. K ROLL , H., E VANGELAKAKIS , C. & V OLL , G. 1993. Two-feldspar geothermometry: a review and revision for slowly cooled rocks. Contributions to Mineralogy and Petrology, 144, 510 –518. M ORIMOTO , T., S ANTOSH , M., T SUNOGAE , T. & Y OSHIMURA , Y. 2004. Spinel þ quartz association from the Kerala khondalites, southern India: evidence for ultrahigh-temperature metamorphism. Journal of Mineralogical and Petrological Sciences, 99, 257–278. M OTOYOSHI , Y. 1986. Prograde and progressive metamorphism of the granulite-facies Lu¨tzow-Holm Bay region, East Antarctica. DSc thesis, Hokkaido University, Sapporo, Japan. M OTOYOSHI , Y. 1998. Ultra-high temperature metamorphism of the Napier Complex, East Antarctica: a metamorphic perspective. Journal of Geological Society of Japan, 104, 794–807 [in Japanese with English abstract]. M OTOYOSHI , Y. & H ENSEN , B. J. 2001. F-rich phlogopite stability in ultrahigh-temperature metapelites from the Napier Complex, East Antarctica. American Mineralogist, 86, 1404–1413. M OTOYOSHI , Y. & I SHIKAWA , M. 1997. Metamorphic and structural evolution of granulites from Rundva˚gshetta, Lu¨tzow-Holm Bay, East Antarctica. In: R ICCI , C. A. (ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antarctica, Siena, 65–72. M OTOYOSHI , Y., M ATSUBARA , S., M ATSUEDA , H. & M ATSUMOTO , Y. 1985. Garnet–sillimanite gneiss from the Lu¨tzow-Holm Complex, East Antarctica. Memoirs of National Institute for Polar Research, Special Issue, 37, 82–94. M OTOYOSHI , Y., M ATSUBARA , S. & M ATSUEDA , H. 1989. P –T evolution of the granulite-facies rocks of the Lu¨tzow-Holm Bay region, East Antarctica. In: D ALY , J. S., C LIFF , R. A. & Y ARDLEY , B. W. D (eds) Evolution of Metamorphic Belts. Geological Society, London, Special Publications, 43, 325 –329. O SANAI , Y., S AJEEV , K., O WADA , M., K EHELPANNALA , K. V. W., P RAME , W. K. B., N AKANO , N. & J AYATILEKE , S. 2006. Metamorphic evolution of highpressure and ultrahigh-temperature granulites from the Highland Complex, Sri Lanka. Journal of Asian Earth Sciences, 28, 20–37. R AITH , M., K ARMAKAR , S. & B ROWN , M. 1997. Ultra-high-temperature metamorphism and multistage decompressional evolution of sapphirine granulites from the Palni Hill Ranges, southern India. Journal of Metamorphic Geology, 15, 379– 399. S AJEEV , K. & O SANAI , Y. 2004a. Osumilite and spinel þ quartz from Sri Lanka: Implications for UHT conditions and retrograde P –T path. Journal of Mineralogical and Petrological Sciences, 99, 320 –327. S AJEEV , K. & O SANAI , Y. 2004b. Ultrahigh-temperature metamorphism (1150 8C, 12 kbar) and multistage evolution of Mg– Al-rich granulites from the central Highland Complex, Sri Lanka. Journal of Petrology, 45, 1821–1844.
S AJEEV , K., O SANAI , Y. & S ANTOSH , M. 2004. Ultrahigh-temperature metamorphism followed by two-stage decompression of Grt– Opx–Sill granulites from Ganguvarpatti, Madrai block, southern India. Contributions to Mineralogy and Petrology, 148, 29–46. S AJEEV , K., S ANTOSH , M. & K IM , H. S. 2006. Partial melting and P –T evolution of the Kodaikanal metapelite belt, southern India. Lithos, 92, 465–483. S ANTOSH , M. & S AJEEV , K. 2006. Anticlockwise evolution of ultrahigh-temperature granulites within continental collision zone in southern India. Lithos, 92, 447–464. S HIRAISHI , K., H IROI , Y. & M OTOYOSHI , Y. 1989a. Geological Map of Lu¨tzow-Holm Bay, Antarctica. Antarctic Geological Map Series, Sheet 12. National Institute of Polar Research, Tokyo. S HIRAISHI , K., H IROI , Y. & M OTOYOSHI , Y. 1989b. Geological Map of Prince Olav Coast, Antarctica. Antarctic Geological Map Series, Sheet 13. National Institute of Polar Research, Tokyo. S HIRAISHI , K., H IROI , Y., E LLIS , D. J., F ANNING , M., M OTOYOSHI , Y. & N AKAI , Y. 1992. The first report of a Cambrian orogenic belt in East Antarctica—An ion microprobe study of the Lu¨tzow-Holm Complex. In: Y OSHIDA , Y., K AMINUMA , K. & S HIRAISHI , K. (eds) Recent Progress in Antarctic Earth Science. Terra, Tokyo, 67–73. S HIRAISHI , K., E LLIS , D. J., H IROI , Y., F ANNING , C. M., M OTOYOSHI , Y. & N AKAI , Y. 1994. Cambrian orogenic belt in East Antarctica and Sri Lanka: Implications for Gondwana assembly. Journal of Geology, 102, 47– 65. S PEAR , F. S., K OHN , M. J. & C HENEY , J. T. 1999. P –T paths from anatectic pelites. Contributions to Mineralogy and Petrology, 134, 17–32. T ATEISHI , K., T SUNOGAE , T., S ANTOSH , M. & J ANARDHAN , A. S. 2004. First report of sapphirine þ quartz assemblage from southern India: implications for ultrahigh-temperature metamorphism. Gondwana Research, 7, 899–912. TSUNOGAE, T. & S ANTOSH , M. 2006. Spinel– sapphirine – quartz bearing composite inclusion within garnet from an ultrahigh-temperature pelitic granulite: Implication for metamorphic history and P– T path. Lithos, 92, 524–536. V IELZEUF , D. & H OLLOWAY , J. R. 1988. Experimental determination of the fluid absent melting relation in the pelieic system. Consequences for crustal differentiation. Contributions to Mineralogy and Petrology, 98, 257–276. Y OSHIMURA , Y., M OTOYOSHI , Y., G REW , E. S., M IYAMOTO , T., C ARSON , C. J. & D UNKLEY , D. J. 2000. Ultrahigh-temperature metamorphic rocks from Howard Hills in the Napier Complex, East Antarctica. Polar Geoscience, 13, 60–85. Y OSHIMURA , Y., M OTOYOSHI , Y., M IYAMOTO , T., G REW , E. S., C ARSON , C. J. & D UNKLEY , D. J. 2004. High-grade metamorphic rocks from Skallevikshalsen in the Lu¨tzow-Holm Complex, East Antarctica: metamorphic conditions and possibility of partial melting. Polar Geoscience, 17, 57–87.
Crystal size distribution of garnet in quartzo-feldspathic gneisses from the Lu¨tzow-Holm Complex at Skallen, East Antarctica SAORI GOTO & TAKESHI IKEDA Department of Earth and Planetary Sciences, Graduate School of Science, Kyushu University, 33 Hakozaki, Fukuoka 812-8581, Japan (e-mail:
[email protected]) Abstract: Crystal size distributions (CSDs) of garnet in quartzo-feldspathic gneisses from the Lu¨tzow-Holm Complex at Skallen, East Antarctica, indicate that significant annealing with Ostwald ripening did not take place even though the rocks underwent regional granulite-facies metamorphism with a peak phase of ultrahigh-temperature conditions. Absence of fluid as a result of the complete consumption of hydrous minerals during garnet-forming reactions could restrain intergranular diffusion and inhibit Ostwald ripening. Garnet-poor gneisses have experienced a single nucleation and growth event associated with a continuous garnet-forming reaction. In contrast, garnet-rich gneisses from the same outcrop have undergone additional garnet-forming reactions responsible for multiple stages of nucleation and growth.
Garnet-bearing quartzo-feldspathic gneisses are one of the main constituent lithologies of the Lu¨tzow-Holm Complex at Skallen, East Antarctica (Yoshida et al. 1976; Osanai et al. 2004). This lithology is composed of simple mineral assemblages such as garnet, quartz, plagioclase, K-feldspar with or without sillimanite. Because of this mineralogical simplicity, the quartzo-feldspathic gneisses have been less studied from a petrological point of view. However, they would be the best subject for investigation of crystal size distributions (CSDs) of garnet, for the following reasons. First, the absence of mafic minerals such as amphiboles, pyroxenes and micas in these gneisses suggests that garnet-forming reactions have been completed and retrograde breakdown of garnet to form these mafic minerals did not take place. Therefore, these rocks are expected to preserve garnet CSDs at high temperatures. Second, the quartzo-feldspathic gneisses investigated in this study are thinly layered, which allows us to reveal the variation of CSDs in different layers that underwent the same pressure– temperature–time path. This study describes the CSDs of garnet from five layers of quartzo-feldspathic gneisses occurring in one outcrop at Skallen, and emphasizes the importance of the absence of hydrous minerals to restrain annealing during formation of CSDs.
facies (Hiroi et al. 1983, 1987). The pressure – temperature paths are inferred as being clockwise in a pressure– temperature diagram with nearly isothermal decompression at high pressures (Motoyoshi et al. 1985; Kawasaki et al. 1993). Two stages of deformation were responsible for the major structures formed after the peak metamorphism (Shiraishi et al. 1983; Ikeda & Kawakami 2004; Kawakami & Ikeda 2004a, b). The metamorphic grade at Skallen, the exposure investigated in this study, belongs to the granulite facies, of which metamorphic conditions were estimated based on geothermobarometry to be 820 8C, 4.1 –6.5 kbar (Yoshida & Aikawa 1983), 730 8C, 6.3 kbar (Suzuki 1983) and 810 8C, 7.7–10.8 kbar (Motoyoshi 1986). However, recent studies of the neighbouring exposures suggest much higher grades that may exceed 1000 8C (e.g. Kawasaki et al. 1993; Motoyoshi & Ishikawa 1997; Kawakami & Motoyoshi 2004). Skallen is underlain mainly by pelitic and quartzo-feldspathic gneisses intercalated with thin layers of mafic gneisses (Yoshida et al. 1976; Osanai et al. 2004). Garnet in garnet–sillimanite quartzo-feldspathic gneisses includes kyanite and staurolite, which is indicative of prograde relic minerals (Motoyoshi et al. 1985). This suggests that garnet grew during the prograde pressure –temperature path passing through the transition from kyanite to sillimanite.
Sample descriptions
Mode of occurrence of quartzo-feldspathic gneisses
Geological outline The Lu¨tzow-Holm Complex, East Antarctica, is a Cambrian orogenic belt that increases in grade southwestward from amphibolite facies to granulite
We investigated garnet-bearing quartzo-feldspathic gneisses at one outcrop in the eastern part of Skallen (698400 0200 S, 398260 2300 E). They are composed of thin layers up to several decimetres thick that are
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 391 –399. DOI: 10.1144/SP308.19 0305-8719/08/$15.00 # The Geological Society of London 2008.
392
S. GOTO & T. IKEDA
(dm), standard deviation (s) and skewness (Sk). The standard deviation and skewness are respectively defined as s¼
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1 X (d dm )2 NA
(1)
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi NA (NA 1) m3 NA 2 s3
(2)
and Sk ¼
Fig. 1. Photograph showing compositional layering of quartzo-feldspathic gneisses at Skallen, Lu¨tzow-Holm Complex, East Antarctica.
approximately parallel (Fig. 1). Each layer can be distinguished from adjacent layers by different modal abundance of garnet. The layers are nearly horizontal or dip gently northward (,208). The gently dipping surface of the ground roughly conforms to the surface of the layerings (Fig. 1). We have taken photographs of the layers from a direction approximately perpendicular to the layer surfaces. Eight rectangular domains in the photographs were chosen so that each of the domains covers the surface of a single layer (Fig. 2). In total, five layers were investigated. Three domains (Fig. 2a – c) are from the same layer (layer A), two (Fig. 2d and e) from a second layer (layer B) and the remaining three domains shown in Fig. 2f – h are from three layers, referred to as layer 33, 34 and 30, respectively. The area of the investigated domains ranges from 240 to 770 cm2. Garnet has a spatially homogeneous occurrence within the domain, and no significant clustering was recognized. The garnet grains are approximately spherical and separated from each other, in contrast to those described by Okamoto & Michibayashi (2005), who found elongated garnet grains at Skallen. It is clear that the grain size and areal fraction of garnet differ in different layers.
Crystal size measurement We measured the area of single garnet grains, A, by counting the number of pixel in the digitalized photographs of Figure 2. The size of garnet grain, d, is defined as the diameter of a circle of equal area to the area occupied by the garnet; that is, d ¼ 2(A/p)0.5. Table 1 summarizes the number of measured garnet grain (NA), number per unit area (Dg), areal fraction of garnet ( f ), mean diameter
where m3 denotes the third moment about dm, defined as m3 ¼
1 X (d dm )3 : NA
(3)
The values of these parameters from three domains in layer A are similar to each other, as is the case for the two domains in layer B. This indicates that a single layer is relatively uniform with respect to CSDs of garnet. In contrast, the areal fraction of garnet, f, varies significantly, from ,10% to .30%, between the layers, and increases with increasing mean diameter, dm, from 2.3 to 4.9 mm (Fig. 3). The number density, Dg, does not vary much between the layers. These features suggest that the layering of the quartzofeldspathic gneisses may be ascribed to a compositional layering with different fraction (modal abundance) and similar number density of garnet. The layers with large f (layers A and 34) have small skewness as compared with those with small f (layers B, 30 and 33) (Fig. 3). Figure 4 shows histograms of garnet grain size divided into 1mm bin size. All the CSDs have a tail in the region larger than the mean grain size, which corresponds to the positive skewness shown in Table 1. The small skewness of layer A reflects the features of relatively short tails and dm belonging to the size class with maximum frequency or its neighbour. In contrast, layer 34, which also represents a small skewness, has the size class with maximum frequency that is smaller than dm and has an irregular tail with large frequency, which results in a large standard deviation. Large skewness for other layers is ascribed to their long tail.
Three-dimensional crystal size distribution The diameter of a grain in a 2D section, d, is equal to or smaller than the true diameter of the grain, L. The 3D CSDs would be different from those shown in Figure 4, in which the number of large
CRYSTAL SIZE DISTRIBUTION OF GARNET
Fig. 2. Photographs of the analysed layer surfaces of quartzo-feldspathic gneisses. (a– c) were taken in different domains of a single layer (layer A); (d) and (e) are from another layer (layer B); (f–h) are from different layers (layers 33, 34 and 30, respectively). Width of field of view: (a) 24.4 cm; (b) 25.1 cm; (c) 27.6 cm; (d) 25.0 cm; (e) 15.6 cm; (f) 25.4 cm; (g) 24.5 cm; (h) 58.7 cm.
393
394
S. GOTO & T. IKEDA
Table 1. Summary of measurements Layer: Domain: Photograph:
A A-1 Fig. 2a
A A-2 Fig. 2b
A A-3 Fig. 2c
B B-1 Fig. 2d
B B-2 Fig. 2e
33 33 Fig. 2f
34 34 Fig. 2g
30 30 Fig. 2h
Measured area (mm2) NA Dg (cm22) f (%) dm (mm) s Sk
42570 848 1.99 23.6 3.61 1.43 0.24
48642 761 1.56 20.0 3.77 1.43 0.34
49225 1132 2.30 21.0 3.15 1.31 0.20
24000 386 1.61 8.2 2.33 1.05 0.83
24925 305 1.22 8.7 2.64 1.46 1.21
48129 632 1.31 7.4 2.52 0.89 0.61
37440 493 1.32 34.3 4.87 3.08 0.31
76607 565 0.74 7.2 3.02 1.83 1.07
NA, number of grain; Dg, number per unit area; f, areal fraction; dm, mean diameter; s, standard deviation; Sk, skewness.
grains increases whereas that of small grains decreases. The positively skewed CSDs in two dimensions may be, therefore, possibly accounted for by the effect of transformation to two dimensions of symmetrical or negatively skewed 3D CSDs. We employed the method of Saltykov (1967) and its modifications (e.g. Higgins 1994, 2000; Kaneko et al. 2005) to estimate the 3D CSDs based on the 2D measurements. Assuming the shape of garnet grain to be spherical, the 3D number density of grains belonging to size between i(d 2 Dd) and id, which is denoted as nvi, was estimated as nvi ¼ M1 naj
(4)
where naj refers to the 2D number density of grains with diameter between j(d 2 Dd) and jd, which was obtained from the measurement. M 21 denotes the inverse matrix of M of which the elements, pij, are expressed as pij ¼ Dd
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffiffiffiffiffi i2 ( j 1)2 i2 j2
for i j and 0 for i , j:
Fig. 3. Results of measurement in terms of mean diameter of garnet, dm, and skewness of size distribution, Sk, v. modal abundance of garnet, f, in areal per cent.
(5)
Dividing pij by L, which is equal to iDd, provides the probability of a sphere with diameter L representing a 2D circle with diameter between j(d 2 Dd ) and jd. The filled circles in Figure 5 show the frequency of garnet grains with 3D diameter L. The 3D CSDs represent asymmetric distributions with long tails on the large-size side. This feature is similar to the 2D CSDs in Figure 4, suggesting that the positive skewness in 2D CSDs reflects 3D CSDs that are positively skewed. The irregular tail in 2D CSDs of layer 34 becomes more significant in 3D CSDs such that the tail is composed of two peaks (Fig. 5). Two peaks are recognized in 3D CSDs
CRYSTAL SIZE DISTRIBUTION OF GARNET
395
Fig. 4. Histograms of number of grains in the measured area, NA, plotted against diameter of grains, d, with 1 mm bin size class. (Note the variation in vertical scale between histograms.)
at maximum frequency in domains A-1 and A-3. Another peak also appears in the relatively flat 2D tail of layer 30.
Discussion Effect of annealing The absence of mafic minerals except garnet in the investigated gneisses suggests that the garnetforming reactions have been completed and the retrograde breakdown of garnet was not significant.
Therefore, the CSDs formed at the nucleation and growth stage would be either preserved or modified to some extent afterwards during an annealing stage. The Ostwald ripening that takes place during an annealing stage tends to coarsen larger grains at the expense of smaller grains to reduce the total surface area of garnet grains (LSW theory: Lifshitz & Slyozov 1961; Wanger 1961), and has been commonly recognized in regional metamorphic terranes (e.g. Cashman & Ferry 1988; Miyazaki 1991). This process results in characteristic CSDs in which the most frequent
396
S. GOTO & T. IKEDA
Fig. 5. Frequency of estimated 3D crystal size. Continuous line represents the best-fit line based on a lognormal distribution function using equation (6).
grain size is larger than dm and the region of grain size smaller than dm has long tails (e.g. Joesten 1991; Miyazaki 1994). The theoretical 3D CSDs can be converted to 2D CSDs that also have the same feature (Miyazaki 1991). This feature is distinctly different from that in Figure 4, suggesting that garnet of the present study did not undergo significant Ostwald ripening. Miyazaki (1991) showed that garnet grains with spatially heterogeneous distribution represent CSDs that are not consistent with those expected by the LSW theory, whereas the CSDs of uniformly distributed grains from the same metamorphic grade are similar to the theoretical CSDs. This suggests that the spatial distribution of grains controls the CSD patterns. However, the spatial distribution of
garnet in this study is roughly uniform (Fig. 2). The spatial distribution could not account for the difference between the measured CSDs in this study and that of Ostwald ripening. The absence of significant evidence for Ostwald ripening is puzzling for rocks that underwent granulite-facies metamorphism because high temperatures would enhance intergranular diffusion, through which Ostwald ripening proceeds. Motoyoshi et al. (1985) noted that the prograde relic minerals were preserved as inclusions in garnet, as described above. This feature may be indicative of high temperatures of relatively short duration and rapid cooling so that there was insufficient time for breakdown of the relic minerals. A rapid cooling history has been verified by
CRYSTAL SIZE DISTRIBUTION OF GARNET
Fraser et al. (2000): it took 17 Ma to cool the rocks undergoing the peak metamorphic condition over 900 8C to 300 –350 8C at Rundva˚gshetta, the highest grade exposure of the Lu¨tzow-Holm Complex. However, retrograde reactions are significant in the biotite-bearing rocks, in which garnet and biotite once equilibrated at high temperatures have re-equilibrated locally during retrograde metamorphism (e.g. Ikeda 2004). This suggests that the high rate of cooling is not the only factor responsible for restraining intergranular diffusion in all the lithological types. The investigated layers do not contain hydrous minerals such as micas and amphiboles. Also, they do not present significant evidence indicative of the presence of melt, such as migmatitic structures. It is, therefore, likely that the rocks were under fluid-absent conditions when the garnetforming reactions were terminated as a result of the complete consumption of nutrient hydrous minerals. The presence of intergranular fluid is essential for Ostwald ripening: its driving force is the difference of chemical potential present in fluid adjacent to grains with different diameters (see Joesten 1991). The exhaustion of intergranular fluid would, therefore, inhibit Ostwald ripening.
Implication for nucleation and growth history Eberl et al. (1998) indicated that CSDs have two basic shapes in addition to the steady-state shape related to Ostwald ripening; that is, asymptotic and lognormal. The 3D CSDs of all the investigated domains show that the frequency in the lowest size class is virtually zero, except for layer 34, which has a maximum frequency in the lowest size class. The slope of the layer 34 CSD, however, becomes gentle with decreasing size (Fig. 5). These features suggest that the investigated CSDs cannot be regarded as being asymptotic. To test whether the CSDs in Figure 5 are conformable to a theoretical lognormal distribution or not, we examined the chi-squared test, as described below. The lognormal distributions are described by the equation 1 (ln L X)2 pffiffiffiffiffiffi exp (6) G(L) ¼ 2b2 bL 2p where X and b2 denote the mean and the standard deviation of the logarithm of grain size, respectively (Eberl et al. 1998). We performed successive approximation to yield X and b2 that provide the minimum reduced chi squared value, e x2 , defined as e x2 ¼
x2 b
(7)
397
where b refers to the statistical degree of freedom, and x2 is defined as ( ) 2 . X .X 2 G(L) : (8) nvi nvi G(L) x ¼ i
i
The best-fitted lognormal distributions are shown as continuous lines in Figure 5. The chisquared test indicates that CSDs of domains B-1 and B-2 are regarded as lognormal at high significance levels (.10 to .20%), and that the CSD of layer 33 has a significance level of 5%. Eberl et al. (1998) indicated that growth at a rate proportional to grain size (law of proportionate effect: LPE) produces a lognormal distribution. This model has accounted for experimental and natural CSDs (Kile et al. 2000; Eberl et al. 2002). Kile et al. (2000) experimentally produced calcite grains representing lognormal CSDs by continuous addition of nutrient solution, which led to a single nucleation event followed by LPE growth. By taking account of the feature that garnet is a solid solution, in contrast to calcite with a fixed composition, the continuous, but not stepwise, supply of nutrient in the experiments may correspond to progress of a single continuous garnet-forming reaction in the present study. This suggests that the continuous garnet-forming reaction was responsible for formation of garnet in layers B and 33, associated with a single nucleation event followed by continuous growth in accordance with the LPE. The chi-squared test indicates that the CSDs of layers A, 34 and 30 are not regarded as lognormal distributions. They have two or three peaks, which suggests that they are composed of several unimodal distributions. These CSDs could be decomposed to two or three lognormal distributions, although the choice of the type of unimodal distribution is arbitrary unless supported by another evidence. This may be indicative of multiple stages of nucleation and growth. Ikeda & Kawakami (2004) recognized two stages of deformation responsible for outcrop pattern and microstructures in the matrix. These deformation stages therefore postdate the formation of garnet and cannot correlate with the inferred multiple stages of nucleation and growth. The fraction of garnet is large (.30%) in layers A and 34, which represent CSDs with several distinguishable peaks (Table 1 and Fig. 5). In contrast, the garnet fraction is small (,10%) in the other layers, which show either lognormal CSDs (layers B and 33) or a nearly lognormal CSD with a second small peak (layer 30). This feature, combined with the above argument, indicates that the garnet-poor layers experienced a continuous garnetforming reaction that caused a single nucleation and growth event, whereas the garnet-rich layers
398
S. GOTO & T. IKEDA
underwent multiple stages of nucleation and growth. In general, several continuous and discontinuous reactions produce garnet with increasing temperature, especially for rocks with bulk chemistry enriched in garnet-constituent components (e.g. Thompson 1976). It would be likely that garnet grains in the garnet-rich layers (layers A and 34) were formed by several garnet-forming reactions. These reactions would have caused multiple stages of nucleation and growth, leading to the irregular CSDs in the garnet-rich layers. Because all the layers occur in the same outcrop and are composed approximately of the same minerals, they would undergo the same garnet-forming reaction through which the nutrient mafic minerals were completely consumed during the same pressure–temperature –time path. This reaction is responsible for the single nucleation and growth event in the garnet-poor layers and also accounts for the final nucleation and growth event occurring in the garnet-rich layers. Several garnet-forming reactions prior to this reaction have caused other nucleation and growth events in the latter layers.
Conclusions The arguments lead to the following conclusions. The CSDs of garnet show asymmetric patterns that have tails on the large-size side, some of which represent lognormal distributions, suggesting that significant annealing with Ostwald ripening did not occur. The investigated gneisses are free of hydrous minerals, which is indicative of fluid-absent conditions caused by the complete consumption of hydrous minerals during garnetforming reactions. The absence of intergranular fluid would restrain intergranular diffusion and therefore inhibit Ostwald ripening. The garnet-poor layers represent lognormal CSDs, in contrast to the garnet-rich layers, which show irregular CSDs. These features can be explained by the former layers having experienced a single nucleation and growth event associated with a continuous garnet-forming reaction, and the latter layers having previously undergone additional garnet-forming reactions responsible for multiple stages of nucleation and growth. The photographs examined in this study were taken during the 44th Japanese Antarctic Research Expedition (JARE) in 2002–2003. We are grateful to all crew members of the icebreaker Shirase and all members of JARE, especially T. Kawakami, Y. Kawano and T. Kawasaki, for their support in performing field research. K. Miyazaki is gratefully acknowledged for his suggestions. Thanks are also due to D. D. Eberl and A. Okamoto for their constructive comments that helped improve the manuscript.
References C ASHMAN , K. V. & F ERRY , J. M. 1988. Crystal size distribution (CSD) in rocks and the kinetics and dynamics of crystallization III. Metamorphic crystallization. Contributions to Mineralogy and Petrology, 99, 401–415. E BERL , D. D., D RITS , V. A. & S RODON , J. 1998. Deducing growth mechanisms for minerals from the shapes of crystal size distributions. American Journal of Science, 298, 499– 533. E BERL , D. D., K ILE , D. E. & D RITS , V. A. 2002. On geological interpretations of crystal size distributions: Constant vs. proportionate growth. American Mineralogist, 87, 1235–1241. F RASER , G., M C D OUGALL , I., E LLIS , D. J. & W ILLIAMS , I. S. 2000. Timing and rate of isothermal decompression in Pan-African granulites from Rundva˚gshetta, East Antarctica. Journal of Metamorphic Geology, 18, 441– 454. H IGGINS , M. D. 1994. Numerical modeling of crystal shapes in thin section: Estimation of crystal habit and true size. American Mineralogist, 79, 113 –119. H IGGINS , M. D. 2000. Measurement of crystal size distributions. American Mineralogist, 85, 1105–1116. H IROI , Y., S HIRAISHI , K., Y ANAI , K. & K IZAKI , K. 1983. Aluminum silicates in the Prince Olav and Soˆya Coasts, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 28, 115–131. H IROI , Y., S HIRAISHI , K., M OTOYOSHI , Y. & K ATSUSHIMA , T. 1987. Progressive metamorphism of calc-silicate rocks from the Prince Olav and Soˆya Coasts, East Antarctica. Proceedings of NIPR Symposium on Antarctic Geosciences, 1, 73–97. I KEDA , T. 2004. Garnet–biotite geothermometry of a pelitic gneiss from the Lu¨tzow-Holm Complex in Skallen, East Antarctica: Constraints on retrograde metamorphism. Polar Geoscience, 17, 45– 56. I KEDA , T. & K AWAKAMI , T. 2004. Structural analysis of the Lu¨tzow-Holm Complex in Akarui Point, East Antarctica, and overview of the complex. Polar Geoscience, 17, 22–34. J OESTEN , R. L. 1991. Kinetics of coarsening and diffusioncontrolled mineral growth. In: K ERRICK , D. M. (ed.) Contact Metamorphism. Mineralogical Society of America, Reviews in Mineralogy, 26, 507–582. K ANEKO , Y., T SUNOGAE , T. & M IYANO , T. 2005. Crystal-size distributions of garnets in metapelites from the northeastern Bushveld contact aureole, South Africa. American Mineralogist, 90, 1422–1433. K AWAKAMI , T. & I KEDA , T. 2004a. Timing of ductile deformation and peak metamorphism in Skallevikshalsen, Lu¨tzow-Holm Complex, East Antarctica. Polar Geoscience, 17, 1– 11. K AWAKAMI , T. & I KEDA , T. 2004b. Structural evolution of the Ongul Islands, Lu¨tzow-Holm Complex, East Antarctica. Polar Geoscience, 17, 12–21. K AWAKAMI , T. & M OTOYOSHI , Y. 2004. Timing of attainment of spinel þ quartz coexistence in the garnet–sillimanite leucogneiss from Skallevikshalsen, Lu¨tzow-Holm Complex, East Antarctica. Journal of Mineralogical and Petrological Sciences, 99, 311–319.
CRYSTAL SIZE DISTRIBUTION OF GARNET K AWASAKI , T., I SHIKAWA , M. & M OTOYOSHI , Y. 1993. A preliminary report on cordierite-bearing assemblages from Rundva˚gshetta, Lu¨tzow-Holm Bay, East Antarctica: evidence for a decompressional P –T path? Proceedings of NIPR Symposium on Antarctic Geosciences, 6, 47– 56. K ILE , D. E., E BERL , D. D., H OCH , A. R. & R EDDY , M. M. 2000. An assessment of calcite crystal growth mechanisms based on crystal size distribution. Geochimica et Cosmochimica Acta, 64, 2937–2950. L IFSHITZ , I. M. & S LYOZOV , V. V. 1961. The kinetics of precipitation from supersaturated solid solution. Journal of Physics and Chemistry of Solids, 19, 35– 50. M IYAZAKI , K. 1991. Ostwald ripening of garnet in high P/T metamorphic rocks. Contributions to Mineralogy and Petrology, 108, 118–128. M IYAZAKI , K. 1994. Ostwald ripening and crystal size distribution (CSD) of garnet in various metamorphic rocks. Proceedings of 29th International Geological Congress, Part A, 27– 38. M OTOYOSHI , Y. 1986. Prograde and progressive metamorphism of the granulite facies Lu¨tzow-Holm Bay region, East Antarctica. Doctoral thesis, Hokkaido University, Sapporo, Japan. M OTOYOSHI , Y. & I SHIKAWA , M. 1997. Metamorphic and structural evolution of granulites from Rundva˚gshetta, Lu¨tzow-Holm Bay, East Antarctica. In: R ICCI , C. A. (ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antartica, Siena, 65– 72. M OTOYOSHI , Y., M ATSUBARA , S., M ATSUEDA , H. & M ATSUMOTO , Y. 1985. Garnet–sillimanite gneisses from the Lu¨tzow-Holm Bay region, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 37, 82– 94. O KAMOTO , A. & M ICHIBAYASHI , K. 2005. Progressive shape evolution of a mineral inclusion under
399
differential stress at high temperature: Example of garnet inclusions within a granulite-facies quartzite from the Lu¨tzow-Holm Complex, East Antarctica. Journal of Geophysical Research, 110, B11203. O SANAI , Y., T OYOSHIMA , T., O WADA , M. ET AL . 2004. Geological map of Skallen, Antarctica (Revised Edition). Antarctic Geological Map Series, Sheet 39. National Institute of Polar Research, Tokyo. S ALTYKOV , S. A. 1967. The determination of the size distribution of particles in an opaque material from a measurement of the size distribution of their sections. In: E LIAS , H. (ed.) Stereology. Proceedings of the Second International Congress for Stereology. Springer, New York, 163– 173. S HIRAISHI , K., H IROI , Y., S ASAKI , K., Y ANAI , K. & K IZAKI , K. 1983. Geological structure of the Prince Olav Coast. Proceedings of NIPR Symposium on Antarctic Geosciences, Abstracts, 79– 80 [in Japanese]. S UZUKI , M. 1983. Preliminary note on the metamorphic conditions around Lu¨tzow-Holm Bay, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 28, 132–143. T HOMPSON , A. B. 1976. Mineral reactions in pelitic rocks: I. Prediction of P– T –X(Fe– Mg) phase relations. American Journal of Science, 276, 401–424. W ANGER , C. 1961. Theorie der Alterung von Niederschla¨gen durch Umlo¨sen (Ostwald-Reifung). Zeitshrift fu¨r Elektrochemie, 65, 581–591. Y OSHIDA , M. & A IKAWA , N. 1983. Petrography of a discordant metabasite from Skallen, Lu¨tzow-Holmbukta, East Antarctica. Memoirs of National Institute of Polar Research, Special Issue, 28, 144–165. Y OSHIDA , M., Y OSHIDA , Y., A NDO , H., I SHIKAWA , T. & T ATSUMI , T. 1976. Geological map of Skallen, Antarctica. Antarctic Geological Map Series, Sheet 9. National Institute of Polar Research, Tokyo.
Contrasting metamorphic P –T path between Schirmacher Hills and Mu¨hlig-Hofmannfjella, central Dronning Maud Land, East Antarctica S. BABA1, M. OWADA2 & K. SHIRAISHI3,4 1
Department of Natural Environment, University of the Ryukyus, 1 Senbaru, Nishihara, Okinawa 903-0213, Japan (e-mail:
[email protected]) 2
Department of Earth Sciences, Yamaguchi University, 1677-1 Yoshida, Yamaguchi 753, Japan
3
Department of Crustal Studies, National Institute of Polar Research, 1-9-10 Kaga, Itabashi-ku, Tokyo 173-8515, Japan
4
Deparment of Polar Science, Graduate School of Advanced Studies (SOKENDAI), 1-9-10 Kaga, Itabashi-ku, Tokyo 173-8515, Japan Abstract: Retrograde metamorphic P– T paths of garnet–pyroxene-bearing mafic gneisses from three regions in central Dronning Maud Land (CDML) were examined. No difference in P– T conditions estimated from rocks of the three regions was recognized, and they are within the range of c. 6– 8 kbar, 750– 830 8C. However, localities in the Mu¨hlig-Hofmann Range (Filchnerfjella and Jutulsessen) preserve rocks with mineral textures that indicate near-isothermal decompressional histories. In the Schirmacher Hills, an isolated exposure on the Princess Astrid Coast, metamorphic texture observed in mafic gneiss is indicative of an isobaric cooling history. Combining their P– T paths and age determinations suggests that the Schirmacher Hills was a separate terrane, together with present-day SE Africa, whereas the Grenvillian-age east– west-trending CDML inland nunatak regions are characterized by an isothermal decompressional metamorphic history related to the final amalgamation of Gondwana.
Central Dronning Maud Land (CDML) has been considered to be an important region for understanding Gondwana and Rodinia supercontinental reconstructions. In the plate-tectonic reconstruction of Gondwana, the DML mountains are inferred to represent the southeastern continuation of the East African Orogen (e.g. Stern 1994; Jacobs et al. 1998). Recent age determinations in the CDML using the U – Pb isotopic signature from zircon in high-grade gneiss rocks gives valuable information with regard to regional correlations of this Neoproterozoic mobile belt with parts of Africa, Madagascar and India (e.g. Jacobs et al. 1998). However, little attention has been paid to the P – T history and its timing, which is critical to improving the understanding of terrane correlations. During the austral summer of 2001 – 2002, the Japanese – Norwegian – German joint geological expedition visited the nunataks of Mu¨hligHofmannfjella and western Orvinfjella in Dronning Maud Land. The expedition also examined exposures in the Schirmacher Hills, an isolated coastal exposure on the Princess Astrid Coast.
In this paper, we compare and contrast the metamorphic conditions of mafic gneisses from the Schirmacher Hills, and two inland regions of CDML (Jutulsessen and Filchnerfjella), and then discuss the tectonic implications of the metamorphic constraints. All mineral abbreviations used in this paper are after Kretz (1983).
Geological background The mountain range of CDML is exposed subparallel to, and c. 200–250 km inland from the edge of the East Antarctica Ice Sheet (Fig. 1). According to Jacobs et al. (2003a), CDML belongs to the East African–Antarctic Orogen (EAAO), which formed as a result of collision of East and West Gondwana during the Pan-African orogeny in late Neoproterozoic to early Palaeozoic time. The older, Mesoproterozoic (c. 1.1 Ga) rocks in CDML were reworked differentially during the PanAfrican orogeny. Metamorphic rocks in Orvinfjella and Wohlthatmassiv (8–148E), eastern CDML, underwent granulite-facies metamorphism between
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 401 –417. DOI: 10.1144/SP308.20 0305-8719/08/$15.00 # The Geological Society of London 2008.
402
S. BABA ET AL.
20°E
40°E
70°
SØ
-
75°
WM 0° .
MH
80°
Fig. 1. Location of Schirmacher Hills, Jutulsessen and Filchnerfjella. SØ, Sør Rondane; WM, Wohlthatmassiv; OV, Orvinfjella; MH, Mu¨hlig-Hofmannfjella; SF, H.U. Sverdrupfjella; HF, Heimefrontfjella.
570 and 530 Ma (Jacobs et al. 2003b). They also show collisional structure and metamorphic zircon growth at c. 580–560 Ma, and large-scale extensional structure associated with post-tectonic intrusion of voluminous granitoid at c. 530–510 Ma. The Pan-African metamorphism and reworking were not intense in the western part of CDML. Voluminous granitoids are exposed over a large area, from H.U. Sverdrupfjella (28E) in the west to Sør Rondane (288E) in the east. The c. 500 Ma granitoids are generally undeformed, except for a few, localized shear zones. Schirmacher Hills is located on the Princess Astrid Coast of CDML (118200 – 118550 E, 708450 S). The metamorphic rocks in the Schirmacher Hills are divided into the following lithologies from the lower to upper structural level: quartzofeldspathic gneiss (including charnockite, mafic granulite), augen gneiss, mixed zone (pelitic gneiss, calcsilicate, mafic granulite, charnockite, etc.), garnet – biotite gneiss, and biotite and biotite – hornblende gneiss (Fig. 2a) (Sengupta 1993). These rocks are interpreted to have
experienced early granulite-facies metamorphism at c. 1150 Ma, followed by amphibolite-facies metamorphism at c. 650 Ma (Grew & Manton 1983). A recent Sm – Nd mineral isochron using orthopyroxene, retrograde garnet reaction rim, and plagioclase gave an age of c. 630 Ma, which is interpreted as the age of cooling from granulitefacies metamorphism (Ravikant et al. 2004). Baba et al. (2006) reported a new sapphirine locality in the southwestern part of the Schirmacher Hills, and proposed ultrahigh-temperature (UHT) metamorphism on the basis of the orthopyroxene compositions and paragenesis. Jutulsessen is located in the central portion of Gjelsvikfjella (28300 – 28500 E, 728000 – 728050 S). A regional geological map of this region (Fig. 2b) has been published by Ohta (1999), who reported that Jutulsessen consists of banded gneisses characterized by alternating felsic (plagioclase, K-feldspar, quartz) and mafic (biotite and amphibole) layers and heterogeneous migmatite containing continuous layeres of amphibolite and micaceous gneiss (Ohta 1999; Paulsson & Austrheim 2003). Investigation near Troll Station showed the presence of garnet – sillimanite-bearing pelitic gneiss and garnet – clinopyroxene-, garnet – hornblende- and garnet – orthopyroxene-bearing mafic gneisses. Most rocks occur as loose scree or talus. Secondary ionization mass spectrometry (SIMS) zircon U – Pb isotopic data suggest a c. 1160 Ma crystallization age for the migmatite protolith, a c. 500 Ma age for the migmatization and a 500 Ma age of emplacement of syenite at Jutulsessen (Paulsson & Austrheim 2003). Bisnath et al. (2006) proposed a two-stage collision event: arc – continent collision with subsequent metamorphism at 1090– 1030 Ma and continent –continent collision at 570 Ma. Filchnerfjella is located near the east edge of the Mu¨hlig-Hofmannfjella, which has near-continuous exposures between 28E and 88E in the CDML. Filchnerfjella is underlain by a series of posttectonic granitoid igneous rocks, which are emplaced into high-grade metamorphic basement rocks (Austrheim et al. 1997; Ohta 1999; Owada et al. 2003; Engvik & Elvevold 2004). The metamorphic rocks in Filchnerfjella are divided into three units; from lowest to highest these are: (1) garnet-bearing leucocratic gneiss (leucogneiss unit); (2) orthopyroxene-bearing brown gneiss (brown gneiss unit); (3) varicoloured-layered gneiss (layered gneiss unit) (Owada et al. 2003). A syenite stock intrudes the metamorphic rocks in the western part of Filchnerfjella, and quartzdioritic to granitic dykes are scattered around Filchnerfjella. Reliable age data have not been obtained yet.
CONTRASTING METAMORPHIC P– T PATH IN CDML (a)
11°40'E
11°30' E
403 11°50'E
Schirmacher Hills
70°45'S
Quartzo-feldspathic gneiss Augen gneiss Mixed zone (metapelites/calc-silicates/ mafic granulites/charnockites) Grt-Bt gneiss Banded gneiss (with px-granulite, charnockite, Leucogneiss quartzofeldspathic gneiss)
Novo GC-1 01120701B 0
2 km
(b) Jutulsessen Banded gneiss Homogeneous migmatite Layered micaceous gneiss
500 ± 8 Ma (SIMS U–Pb) Intrusion age
Amphibolite Gabbro Syenite
72°S 504 ± 6 Ma (SIMS U–Pb) Migmatisation
Troll Grt-M2 Troll Grt-M5
0
5 km 2°40'E
(c)
1163 ± 6 Ma (SIMS U–Pb) Zircon core
7°30'E
Layered gneiss unit Brown gneiss unit
Filchnerfjella
Leucogneiss unit Syenite
02010201B
Antiform Fault 02010601B, E
0
5 km 72°S
Fig. 2. Simplified geological map of study area. (a) Schirmacher Hills (modified after Sengupta 1993; Rameshwar Rao et al. 1997); (b) Jutulsessen (modified after Ohta 1999); (c) Filchnerfjella (Owada et al. 2003). The stars indicate the sample localities in this study.
404
S. BABA ET AL.
Petrography Schirmacher Hills Two mafic gneisses (Novo GC-1 and 01120701B), collected from adjoining localities of sapphirinebearing rocks, have a mineral assemblage of Grt–Cpx– Opx–Qtz –Pl–Opq. Sample Novo GC-1 is a coarse-grained plagioclase-rich (up to 10 mm in diameter) rock composed of plagioclase, garnet, clinopyroxene, quartz, ilmenite, orthopyroxene, and small amount of K-feldspar and secondary biotite. The garnet and clinopyroxene occur as small grains at the margin of coarse-grained plagioclase, and are zoned with garnet in the inner part and clinopyroxene in the outer part (Fig. 3a). Small amounts of orthopyroxene occur as tiny grains in the clinopyroxene aggregate. Quartz occurs within the matrix of garnet and clinopyroxene whereas large irregular-shaped ilmenite grains enclose rounded quartz and clinopyroxene. Sample 01120701B was collected from the mafic-rich gneiss layer and consists of clinopyroxene, garnet, plagioclase, quartz, hornblende, orthopyroxene and cummingtonite, with ilmenite
and apatite in trace amounts. Idioblastic clinopyroxene is enclosed by irregular-shaped garnet that, in places, contains vermicular ilmenite (Fig. 3b). Tiny orthopyroxene grains are also enclosed by garnet. In places, clinopyroxene is pseudomorphically replaced by cummingtonite. Brown and green hornblende are also present; the latter is dominantly developed marginal to clinopyroxene grains.
Jutulsessen Two garnet-bearing mafic gneisses with clinopyroxene and/or orthopyroxene were examined. Sample Troll Grt-M2 is composed of garnet, hornblende, quartz, orthopyroxene, plagioclase and ilmenite. Garnet porphyroblasts are replaced by a coronal symplectite of orthopyroxene and plagioclase (Fig. 4a). These symplectites and coronas are also formed on the brown hornblende margin. Ilmenite occurs in various textural setting. In places, vermicular-shaped ilmenite grains are well developed in orthopyroxene and hornblende. Sample Troll Grt-M5 consists of clinopyroxene, garnet, plagioclase and hornblende, with
Fig. 3. Photomicrographs showing mineral relationships in the garnet–pyroxene-bearing mafic gneiss from the Schirmacher Hills. (a) Garnet (Grt) and clinopyroxene (Cpx) occurrences. Small garnet grains are developed at the marginal part of large plagioclase (Pl) grains. Clinopyroxene and quartz (Qtz) are observed between orthopyroxene (Opx) and plagioclase. (b) Garnet–clinopyroxene –orthopyroxene– quartz association in mafic gneiss. Scale bars represent 2 mm.
CONTRASTING METAMORPHIC P– T PATH IN CDML
405
Fig. 4. Photomicrographs showing mineral relationships in the garnet– pyroxene-bearing mafic gneisses from Jutulsessen. (a) Symplectitic orthopyroxene and plagioclase is developed between garnet and brown hornblende (Hbl) porphyroblast. (b) Garnet–clinopyroxene –quartz association in mafic gneiss (plane- and cross-polarized light). Scale bars represent 2.0 mm.
small amounts of quartz, orthopyroxene, ilmenite and apatite (Fig. 4b). Medium-grained clinopyroxene is abundant throughout the sample, and is locally enclosed by irregular-shaped garnet. Poikilitic plagioclase grains are partly developed, and enclose rounded clinopyroxene, garnet, ilmenite and quartz grains. Hornblende is secondary, partially replacing clinopyroxene.
Filchnerfjella In this area, high-grade rocks with garnet–clinopyroxene association have not been observed. The metamorphic P–T conditions of three garnet–orthopyroxene mafic gneisses were examined. Sample 02010601B was collected from mafic gneiss in the layered gneiss unit and has a mineral
406
S. BABA ET AL.
assemblage of garnet þ hornblende þ orthopyroxene (as symplectite) + spinel þ plagioclase þ quartz associations (garnet pod-type: Owada et al. 2003). It consists of mainly garnet porphyroblast, hornblende, symplectic orthopyroxene (Fig. 5a), plagioclase, biotite and spinel, with trace amounts of quartz. Orthopyroxene occurs only as vermicular
symplectitic crystals together with plagioclase, and replaces the garnet and hornblende margin. Sample 02010601E belongs to the layered-type garnet–orthopyroxene mafic gneiss, having a garnet þ hornblende þ orthopyroxene (fine-grained) þ plagioclase + biotite + quartz association in the layered gneiss unit (Owada et al. 2003). Garnet and
Fig. 5. Photomicrographs showing mineral relationships in the garnet–pyroxene-bearing mafic gneisses from Filchnerfjella. (a) Symplectitic orthopyroxene is developed between garnet and brown hornblende porphyroblast. (b) Mode of occurrence of garnet– orthopyroxene–hornblende association in typical mafic gneiss in this region. Orthopyroxene occurs as tiny grains adjacent to hornblende. Scale bars represent 2.0 mm.
CONTRASTING METAMORPHIC P– T PATH IN CDML
407
hornblende are coarser in grain size than other phases, and orthopyroxene, occurring between those minerals, is subordinate in grain size (Fig. 5b). Sample 02010201B is composed of K-feldspar, quartz, plagioclase, orthopyroxene, hornblende, garnet, ilmenite and fayalite. The rock is homogeneous in nature and belongs to the brown gneiss unit. Large amounts of K-feldspar and quartz are coarser than the other phases. The existence of fine-grained fayalite in the orthopyroxene- and hornblende-rich portions in the matrix is a distinctive feature as compared with other rocks. Further details of the mineral textures of these rocks have been reported by Owada et al. (2003).
in which orthopyroxene occurs as symplectite (02010601B). There is a positive correlation between XMg values of orthopyroxene and coexisting clinopyroxene. Orthopyroxene in 01120701B has higher XMg values (0.43 –0.44) than those in Novo GC-1 (0.33–0.34), and the same pattern occurs for clinopyroxene (Fig. 6). The lowest XMg values are preserved in orthopyroxene in the fayalitebearing sample (02010201B) from Filchnerfjella. As the symplectitic and porphyroblastic orthopyroxene do not occur in the same sample, their compositional differences could not be discussed.
Mineral Chemistry
Clinopyroxene occurs only in samples from the Schirmacher Hills and sample Troll Grt-M5. Clinopyroxene in Novo GC-1 and 01120701B from the Schirmacher Hills shows XMg of 0.45–0.49 and 0.53–0.62, and Al2O3 of 2.0–2.4 wt% and 1.5–3.0 wt%, respectively. In sample 01120701B, clinopyroxene grains adjacent to garnet tend to show high XMg values and low Al2O3 contents (0.60–0.62, 1.5–2.0 wt%). The sample from Jutulsessen (Troll Grt-M5) has XMg of 0.49–0.53 and Al2O3 of 0.7–2.1 wt%. MnO contents of clinopyroxene from sample Troll Grt-M5 are higher (0.60–0.82 wt%) than those for other samples (01120701B: 0.12–0.21 wt%; Novo GC-1: 0.11– 0.23 wt%).
The chemical compositions of constituent minerals were analysed using a wavelength-dispersive electron microprobe (JEOL JXA-8800M) at the National Institute of Polar Research (Japan). Mineral analyses were performed using an accelerating voltage of 15 kV and specimen current of 4–12 nA. Natural minerals and synthetic oxides were used as standards. Mineral compositional data are presented in Tables 1 and 2. Garnet and pyroxenes compositional plots are shown in Figure 6.
Garnet All garnet grains analysed in this study have high almandine and grossular contents. Garnet in samples Novo GC-1 and 01120701B from the Schirmacher Hills have a high almandine content of 0.61– 0.62 and 0.64–0.65, respectively. Except for 02010201B, garnet grains from Jutulsessen and Filchnerfjella have almandine contents within the range of 0.43–0.61, and grossular contents of 0.10– 0.27. Garnet in 02010201B (fayalite-bearing sample) shows high almandine (0.61 –0.63), high grossular (0.19–0.20) and moderately high spessartine (0.14 –0.16) contents. The spessartine content of analysed garnet grains is generally low, except in garnet from sample Troll Grt-M5, which has high spessartine contents (0.22 –0.25). Garnet replaced by orthopyroxene–plagioclase symplectite preserves compositional zoning with a decrease in grossular content toward the rim (e.g. 02010 601B: core 0.17 –0.15 and outermost rim 0.12; Troll Grt-M2: core 0.24 and outermost rim 0.22) and slight increase in spessartine content from the mantle. A decrease in pyrope content is also recognized (e.g. 0.400 for the core to 0.356 for the rim in 02010601B), although it is not obvious in all grains.
Clinopyroxene
Hornblende Hornblende has essentially similar XMg values to those of orthopyroxene, and has low TiO2 contents up to 2.0 wt%.
Plagioclase Matrix plagioclase in the samples has a wide range of XAn values regardless of coexisting garnet grossular content. In the orthopyroxene–plagioclase symplectite, the inner side of plagioclase adjacent to garnet rims shows higher XAn values than those in the outer part. As matrix plagioclase is lacking in both symplectite-bearing samples, the compositional difference cannot be compared. Plagioclase inclusions in garnet also show lower XAn than those in symplectite.
Fayalite Fayalite occurs only in sample 02010201B, and has low XMg of 0.04 –0.06.
Orthopyroxene
Metamorphic P – T conditions
Orthopyroxene grains have low Al2O3 content and moderate XMg values, except for one sample
The P–T conditions of equilibration of the metamorphic mineral associations have been estimated
408
Table 1. Representative chemical compositions of constituent minerals Sample: Analysis no.: Mineral: Note:
O Si Ti Al Cr Fe Mn Mg Ca Na K Total cation XMg Xgrs Xprp Xsps XAn
01120701B
Troll Grt-M2
46 grt
49 cpx
53 opx
42 pl
59 hbl
22 grt
6 cpx
16 opx
7 pl
8 hbl
66 grt core
72 grt rim
69 opx
89 pl cent*
83 pl to grt†
74 hbl
38.50 0.13 21.17 0.00 31.06 0.78 2.74 7.03 0.04 0.00 101.45
50.38 0.26 2.17 0.01 15.77 0.07 8.34 21.59 0.46 0.00 99.05
49.63 0.03 1.06 0.00 38.05 0.23 10.82 0.69 0.03 0.00 100.54
57.09 0.00 27.11 0.00 0.06 0.00 0.01 9.65 6.11 0.27 100.3
36.35 0.03 20.34 0.00 30.48 0.68 2.76 6.97 0.03 0.00 97.64
38.57 0.00 21.51 0.02 29.26 0.96 3.12 7.42 0.04 0.00 100.9
51.14 0.34 2.81 0.02 14.54 0.12 9.38 21.34 0.51 0.00 100.2
50.40 0.08 0.83 0.00 32.59 0.31 14.35 0.55 0.08 0.00 99.19
51.32 0.01 31.56 0.00 0.00 0.00 0.00 13.95 3.19 0.08 100.11
41.26 2.01 12.30 0.02 19.06 0.00 7.22 11.62 1.32 1.99 96.80
39.05 0.04 22.13 0.00 26.31 0.75 4.29 8.90 0.02 0.00 101.49
38.87 0.02 22.42 0.00 26.58 0.59 4.70 8.16 0.03 0.00 101.37
50.96 0.09 0.51 0.00 34.25 0.46 13.67 0.51 0.03 0.00 100.48
52.35 0.00 30.50 0.02 0.17 0.00 0.00 13.17 4.03 0.12 100.36
47.39 0.07 34.35 0.00 0.10 0.01 0.01 16.98 1.92 0.02 100.85
44.95 1.65 9.90 0.02 19.36 0.22 9.02 10.49 1.35 0.76 97.72
12 3.02 0.01 1.96 0.00 2.04 0.05 0.32 0.59 0.01 0.00 8.00
6 1.96 0.01 0.10 0.00 0.51 0.00 0.48 0.90 0.03 0.00 4.00
6 1.98 0.00 0.05 0.00 1.27 0.01 0.64 0.03 0.00 0.00 3.99
8 2.56 0.00 1.43 0.00 0.00 0.00 0.00 0.46 0.53 0.02 5.00
23 5.71 0.00 3.77 0.00 4.01 0.09 0.65 1.17 0.01 0.00 15.41
12 3.02 0.00 1.99 0.00 1.92 0.06 0.36 0.62 0.01 0.00 7.99
6 1.95 0.01 0.13 0.00 0.46 0.00 0.53 0.87 0.04 0.00 4.00
6 1.99 0.00 0.04 0.00 1.08 0.01 0.84 0.02 0.01 0.00 3.99
8 2.33 0.00 1.69 0.00 0.00 0.00 0.00 0.68 0.28 0.00 4.97
23 6.37 0.23 2.24 0.00 2.46 0.00 1.66 1.92 0.40 0.39 15.67
12 3.01 0.00 2.01 0.00 1.69 0.05 0.49 0.73 0.00 0.00 7.99
12 2.99 0.00 2.03 0.00 1.71 0.04 0.54 0.67 0.00 0.00 7.99
6 2.00 0.00 0.02 0.00 1.12 0.02 0.80 0.02 0.00 0.00 3.99
8 2.37 0.00 1.63 0.00 0.01 0.00 0.00 0.64 0.35 0.01 5.00
8 2.16 0.00 1.84 0.00 0.00 0.00 0.00 0.83 0.17 0.00 5.00
23 6.77 0.19 1.76 0.00 2.44 0.03 2.02 1.69 0.39 0.15 15.44
0.485
0.336
0.535
0.440
0.136 0.197 0.107 0.053
*Composition of central part of symplectite. † Plagioclase adjacent to garnet.
0.139
0.466
0.160 0.210 0.123 0.060
0.403
0.707
0.225 0.247 0.166 0.038
0.235 0.227 0.182 0.031
0.416
0.454
0.644
0.830
S. BABA ET AL.
SiO2 TiO2 Al2O3 Cr2O3 FeOT MnO MgO CaO Na2O K2O Total
NOVO GC-1
Table 2. Representative chemical compositions of constituent minerals Sample: Analysis no.: Mineral: Note:
O Si Ti Al Cr Fe3þ* Fe2þ Mn Mg Ca Na K Total cation XMg Xgrs Xprp Xsps XAn
28 grt
36 cpx
44 pl
38.54 51.61 46.06 0.04 0.07 0.00 21.22 0.87 35.11 0.03 0.00 0.00 27.78 16.05 0.04 4.29 0.64 0.00 1.90 9.02 0.00 7.90 21.91 18.42 0.00 0.17 1.12 0.00 0.00 0.02 101.70 100.34 100.77
02010601B 42 hbl
39 grt core
35 grt rim
31 opx symp†
24 pl symp†
02010601E 52 pl in grt‡
44.69 39.75 39.97 51.66 45.18 47.03 1.17 0.00 0.01 0.03 0.01 0.00 9.59 22.52 22.22 3.97 35.38 34.00 0.03 0.00 0.00 0.00 0.00 0.00 20.92 24.90 24.94 22.05 0.18 0.15 0.34 0.63 1.01 0.30 0.00 0.00 7.58 8.69 9.10 22.74 0.01 0.00 11.59 5.22 4.57 0.17 18.34 17.22 1.12 0.00 0.01 0.00 0.92 1.69 0.67 0.00 0.00 0.00 0.00 0.00 97.70 101.71 101.83 100.92 100.02 100.09
12 3.03 0.00 1.96 0.00
6 1.99 0.00 0.04 0.00
8 23 2.10 6.80 0.00 0.13 1.89 1.72 0.00 0.00
12 3.00 0.00 2.00 0.00
12 3.01 0.00 1.97 0.00
1.82 0.29 0.22 0.66 0.00 0.00 7.99
0.52 0.02 0.52 0.90 0.01 0.00 4.00
0.00 2.66 0.00 0.04 0.00 1.72 0.90 1.89 0.10 0.33 0.00 0.13 5.00 15.43
1.57 0.04 0.98 0.42 0.00 0.00 8.00
1.57 0.06 1.02 0.37 0.00 0.00 8.01
0.109 0.222 0.074 0.243
0.500
0.383 0.140 0.325 0.028
0.394 0.122 0.338 0.044
0.392
0.901
8 2.08 0.00 1.92 0.00 0.01 0.00 0.00 0.00 0.91 0.08 0.00 5.00
0.712
0.917
0.849
60 opx
43.29 38.63 50.30 0.71 0.06 0.05 14.41 21.68 1.06 0.03 0.00 0.00 10.87 27.25 34.61 0.19 0.70 0.30 15.08 4.04 14.07 9.82 9.06 0.56 2.54 0.02 0.00 0.53 0.00 0.00 97.47 101.44 100.95
8 23 2.16 6.29 0.00 0.08 1.84 2.47 0.00 0.00 0.01 0.00 1.32 0.00 0.02 0.00 3.27 0.85 1.53 0.15 0.72 0.00 0.10 5.00 15.80
0.662
58 grt
12 2.99 0.00 1.98 0.00 1.77 0.05 0.47 0.75 0.00 0.00 8.01 0.209 0.248 0.154 0.036
6 1.97 0.00 0.05 0.00 0.02 1.11 0.01 0.82 0.02 0.00 0.00 4.00
51 pl 46.80 0.02 33.99 0.00 0.11 0.00 0.00 17.37 1.62 0.00 99.91
54 hbl
1 grt
14 opx
8 pl
41.49 37.39 47.01 59.34 2.06 0.02 0.12 0.00 13.17 20.72 0.35 26.00 0.01 0.00 0.00 0.00 20.45 33.01 47.29 0.09 0.15 1.72 0.96 0.00 8.12 0.88 3.19 0.00 10.33 6.89 0.81 7.62 2.67 0.03 0.03 7.45 0.21 0.00 0.00 0.12 98.66 100.66 99.76 100.62
22 hbl 39.65 1.77 11.04 0.00 29.77 0.20 2.52 10.21 1.62 1.56 98.34
8 23 2.15 6.26 0.00 0.23 1.84 2.34 0.00 0.00
12 3.01 0.00 1.96 0.00
6 2.00 0.00 0.02 0.00
8 23 2.63 6.34 0.00 0.21 1.36 2.08 0.00 0.00
0.00 2.58 0.00 0.02 0.00 1.83 0.86 1.67 0.14 0.78 0.00 0.04 5.00 15.75
2.22 0.12 0.11 0.59 0.00 0.00 8.01
1.68 0.03 0.20 0.04 0.00 0.00 3.99
0.00 3.98 0.00 0.03 0.00 0.60 0.36 1.75 0.64 0.50 0.01 0.32 5.01 15.82
0.424
0.414
0.856
0.045 0.107 0.196 0.035 0.143
0.131
0.361
409
*Fe3þ for orthopyroxene is estimated from charge balance. † Grain occurs as symplectite. ‡ Inclusions in garnet.
6 1.89 0.00 0.17 0.00 0.04 0.64 0.01 1.24 0.01 0.00 0.00 4.00
43 hbl
02010201B
CONTRASTING METAMORPHIC P– T PATH IN CDML
SiO2 TiO2 Al2O3 Cr2O3 FeOT MnO MgO CaO Na2O K2O Total
Troll Grt-M5
410
S. BABA ET AL. 0.6
(a) Schirmacher Hills Ca/(Mg + Ca + Fe)
0.5
Novo GC-1
Cpx
01120701B
0.4 0.3 0.2
Grt
0.1
KD = 3.04
0 0
KD = 4.01
Opx 0.1
0.2
0.3
0.4
0.5
0.6
0.7
Mg/(Mg + Ca + Fe) 0.6
(b) Jutulsessen
Troll Grt-M5
Ca/(Mg + Ca + Fe)
0.5
Troll Grt-M2
Cpx 0.4 0.3
Grt 0.2
KD = 2.63–2.91 0.1
Opx 0 0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Mg/(Mg + Ca + Fe) 0.6
02010601B 02010601E 02010201B
(c) Filchnerfjella Ca/(Mg + Ca + Fe)
0.5
using experimentally calibrated geothermometers based on garnet–orthopyroxene and garnet– clinopyroxene, and geobarometers based on garnet– orthopyroxene/clinopyroxene –plagioclase –quartz equilibria. A compositionally homogeneous core to mantle of orthopyroxene, clinopyroxene and garnet was used for calculation. In the case of the symplectite-bearing assemblage, garnet rim composition and adjacent plagioclase and orthopyroxene were selected (Tables 3 and 4). Fe end-member reaction models were employed for calculations, because all minerals have Fe-rich compositions. Temperatures calculated using garnet– orthopyroxene Fe –Mg exchange thermometry (Lee & Ganguly 1988; Bhattacharya et al. 1991) range from 706 to 964 8C at 7 kbar. These temperature estimates are minima because of possible Fe –Mg re-equilibration during cooling (e.g. Fitzsimons & Harley 1994). Pressures calculated using garnet–orthopyroxene–plagioclase–quartz barometry (Perkins & Chipera 1985; Bhattacharya et al. 1991) at 800 8C are 7.0–7.8 kbar for Schirmacher Hills, 6.3–7.5 kbar for Jutulsessen and 6.2–9.1 kbar for Filchnerfjella. Garnet–clinopyroxene–plagioclase–quartz barometry (Moecher et al. 1988; Eckert et al. 1991) applied to samples from the Schirmacher Hills and Jutulsessen gives 7.7–8.5 kbar at 800 8C and 5.9–6.8 kbar at 800 8C, respectively. Temperatures calculated using garnet–clinopyroxene Fe–Mg exchange thermometry (e.g. Ellis & Green 1979; Ganguly et al. 1996) range from 759 to 789 8C for the Schirmacher Hills. The sample from Jutulsessen gives relatively high T estimates of 779–844 8C, although these are doubtful because of the high spessartine contents in garnet. Figure 7 shows a summary of P–T results for the three regions studied.
0.4
Discussion 0.3
Interpretation of metamorphic textures and conditions
Grt KD = 2.74–3.56
0.2
KD = 2.83 KD = 3.03–3.24
0.1 0 0
Opx 0.1
0.2
0.3
0.4
0.5
0.6
0.7
Mg/(Mg + Ca + Fe)
Fig. 6. Compositional variation of garnet, orthopyroxene and clinopyroxene in terms of Mg/(Mg þ Ca þ Fe) v. Ca/(Mg þ Ca þ Fe). All garnets show low pyrope contents (almost equal to Mg/(Mg þ Ca þ Fe)) and high Ca content in Schirmacher Hills and Jutulsessen. In the Filchnerfjella, the symplectitic orthopyroxene and garnet pair shows high Mg content. Open oval-shaped fields in Filchnerfjella represent the compositional range of those in other garnet–orthopyroxene-bearing gneisses.
The mineral texture of garnet aggregate on the plagioclase margin observed in Novo GC-1 from the Schirmacher Hills (Fig. 3) is inferred to have formed during isobaric cooling (IBC: e.g. Harley 1989). The possible metamorphic reaction is as follows: plagioclase þ orthopyroxene ¼ garnet þ clinopyroxene þ quartz: The P–T conditions obtained from adjacent garnet– orthopyroxene–plagioclase–quartz assemblage and garnet–clinopyroxene–plagioclase–quartz assemblage lie within the range of c. 6–8 kbar, 770– 820 8C, as shown in Figure 7. The proposed reaction
Table 3. Thermobarometric results for Grt –Opx and Grt – Opx – Pl– Qtz associations Sample
Opx
Pl
XMg
Xgrs
Xalm
Xpyp
XMg
XAl(Al/2)
An
0.138 0.193
0.199 0.200
0.677 0.651
0.109 0.127
0.328 0.439
0.027 0.017
0.448 0.702
0.189 0.201
0.211 0.219
0.622 0.613
0.145 0.154
0.405 0.40
0.016 0.012
0.046 0.394 0.170
0.195 0.122 0.180
0.731 0.519 0.660
0.035 0.338 0.136
0.119 0.663 0.423
0.011 0.087 0.023
KD
P (at 800 8C) (kbar)
T (at 7 kbar) (8C) Ha
LG
CH
Bat
PN
PCFe
BatFe
3.04 4.02
728 612
862 729
753 633
814 706
6.9 5.1
7.5 7.8
7.1 7.0
0.830 0.644
2.92 2.63
754 811
895 964
779 839
841 899
6.7 7.9
7.1 7.5
6.3 6.6
0.369 0.913 0.855
2.89 3.03 3.56
763 705 652
901 809 769
789 730 675
816 780 744
6.9 7.1 4.9
6.7 8.8 6.8
6.5 6.3 6.2
KD ¼ (Fe/Mg)Grt/(Fe/Mg)Opx. Ha, Harley (1984); LG, Lee & Ganguly (1988); CH, Carswell & Harley (1989); Bat, Bhattacharya et al. (1991); PN, Perkins & Newton (1981); PCFe, Perkins & Chipera (1985); BatFe, Bhattacharya et al. (1991).
Table 4. Thermobarometric results for Grt –Cpx and Grt – Cpx – Pl– Qtz associations Sample
Schirmacher Hills Novo GC-1 Novo GC-1 01120701B 01120701B Jutulsessen Troll Grt-M5 Troll Grt-M5
Grt
Cpx
Pl
P (at 800 8C) (kbar)
T (at 8 kbar) (8C)
XMg
Xgrs
Xalm
Xpyp
XMg
XAl(Al/2)
An
E&G
G
K
E
MoFe
0.136 0.136 0.162 0.156
0.197 0.197 0.210 0.204
0.643 0.643 0.607 0.622
0.053 0.053 0.058 0.053
0.485 0.485 0.552 0.528
0.050 0.050 0.059 0.065
0.448 0.466 0.706 0.702
767 767 759 763
789 789 777 784
705 705 702 705
7.1 6.9 6.8 6.4
7.8 7.7 8.5 8.0
0.121 0.117
0.222 0.23
0.458 0.463
0.237 0.227
0.497 0.519
0.025 0.054
0.901 0.914
844 809
811 779
747 714
7.9 7.7
5.9 6.8
CONTRASTING METAMORPHIC P– T PATH IN CDML
Schirmacher Hills Novo GC-0.138 01120701B Jutulsessen Troll Grt-M2 Troll Grt-M2 Filchnerfjella 02010201B 02010601B 02010601E
Grt
E&G, Ellis & Green (1979); G, Ganguly et al. (1996); K, Krogh (1988); E, Eckert et al. (1991); MoFe, Moecher et al. (1988).
411
412
S. BABA ET AL.
Fig. 7. P– T plot showing geobarometry applied to garnet– clinopyroxene– plagioclase –quartz (†) and garnet– orthopyroxene–plagioclase– quartz association (W). P– T result deduced from garnet– hornblende– plagioclase –quartz association by Owada et al. (2003) is also shown in Filch 02010601B. Univariant lines were obtained from an experimental study of Qtz-tholeiites having variable mg-number (of 10, 60 and 90) by Green & Ringwood (1967). Abbreviations are as in Tables 3 and 4.
CONTRASTING METAMORPHIC P– T PATH IN CDML
becomes a continuous reaction in the CaO–FeO– MgO–Al2O3 –SiO2 (CFMAS) system, and might progress within the range of inferred P–T conditions. A similar metamorphic texture, garnet corona developed over orthopyroxene supportive of the above reaction, has been proposed by Dasgupta et al. (2001), and they calculated T conditions of 780– 840 8C on the basis of Grt–Cpx association. The temperature estimates obtained from garnet–orthopyroxene–plagioclase–quartz equilibria are broadly consistent with those obtained from clinopyroxenebearing associations. A similar P–T condition was also obtained by fluid inclusion study of mafic gneiss (Rameshwar Rao et al. 1997: c. 837 8C, 7.1 kbar), calc-silicate assemblage (Piazolo & Markl 1999: c. .750 8C, ,8 kbar) and mafic enclaves (Ravikant 2005: ,800 8C, ,8 kbar). It is obvious that the garnet porphyroblast breakdown texture to form orthopyroxene–plagioclase symplectites observed in both Filchnerfjella and Jutulsessen is indicative of near-isothermal decompression (e.g. Harley 1989). The decrease in grossular content at the garnet rim in the above textural setting also supports this interpretation. In addition, orthopyroxene replacing hornblende suggests a slight increase in temperature by dehydration reaction (see Figs 4a and 5a, b). The absence of an obvious decrease in pyrope content of garnet would be caused by the heating. The texture is consistent with the following continuous reaction in the CFMAS system: garnet þ hornblende þ quartz ¼ orthopyroxene þ plagioclase þ vapour: Owada et al. (2003) reported a two-stage metamorphic history from the rocks in Filchnerfjella, where the earlier stage of metamorphism (9 kbar, 600 –700 8C) was defined as a porphyroblast stage (with garnet and hornblende stable), and the later one (7.2 kbar, 830 8C) was recognized as a symplectite stage (with either orthopyroxene or cordierite stable). In metapelitic lithologies cordierite formation is pronounced instead of orthopyroxene formation, indicating a decrease in pressure. As mentioned above, there are no major differences in P– T conditions obtained from garnetbearing mafic gneisses in the three regions (see Fig. 7). However, comparison of mineral textural features suggests that the metamorphic P–T path of the Schirmacher Hills granulites is typical of an IBC history whereas those from Jutulsessen and Filchnerfjella are typical of an isothermal decompression (ITD) history. The IBC history observed in the Schirmacher Hills may represent a cooling path after a period of UHT metamorphism
413
(c. 950– 1050 8C, 9–10 kbar) as proposed by Baba et al. (2006). The pressure conditions obtained from Schirmacher Hills and Jutulsessen seem to be lower in spite of containing the high-pressure indicator Grt –Cpx– Qtz. The appearance of the Grt –Cpx–Qtz –Pl association at a moderate pressure condition (c. up to 8.5 kbar) might be attributed to the Fe-rich whole-rock compositions as reflected by the high Fe contents of garnet, clinopyroxene and orthopyroxene. The appearance of garnet and pyroxenes under relatively low-P condition (e.g. 7 kbar at 800 8C: Mg-number 10) is consistent with the results obtained from the high-pressure experiments on quartz tholeiite (Green & Ringwood 1967).
Regional correlation of metamorphic P –T paths Figure 8 shows metamorphic P–T paths of the highgrade rocks from CDML nunataks and Schirmacher Hills. In Hochlinfjellet, Gjelsvikfjella, located between Jutulsessen and Filchnerfjella, BucherNurminen & Ohta (1993) reported P–T conditions of 750 8C and 8 kbar from orthopyroxene–garnet granulite, and 650 8C and 4 kbar from garnet– cordierite gneiss. They proposed two models: either (1) the two types of gneisses formed as products of different metamorphic events, or (2) the two types of gneisses formed during a single orogenic cycle (Fig. 8, paths 1a and 1b). Our data suggest the latter possibility; that is, two textural developments related to different stages of a single metamorphism, similar to those observed in rocks from Filchnerfjella (Owada et al. 2003). Engvik & Elvevold (2004) proposed an isothermal metamorphic history for Filchnerfjella on the basis of qualitative analyses. Figure 9 shows a microstructure of orthopyroxene symplectite from mafic gneiss collected from their ‘location B’ in the Hochlinfjellet region (Bucher-Nurminen & Ohta 1993). The sample lacks garnet and quartz, although we assume that the symplectitic texture reflects complete consumption of the reactants, and is essentially coeval with those observed in rocks from Jutulsessen and Filchnerfjella. Therefore, the regional distribution of rocks that have undergone metamorphism accompanied by ITD to form orthopyroxene symplectite after garnet is traceable along an east –west trend, parallel to the inland nunatak trend. In addition, Bisnath & Frimmel (2005) suggested a decompressional path for the rocks in the Jutulsessen region (Fig. 8, path 10). Their high-pressure condition was extrapolated from the adjacent H.U. Sverdrupfjella region, and it is not clear whether Jutulsessen experienced such high-P conditions or not, but the
414
S. BABA ET AL.
P (kbar)
P (kbar)
Gjelsvikfjella
14
14 10
Ky
12
12
Schirmacher Hills Ky 5
10
10
8
8
6 4
2
4
Sil
1a
200
400
Sil
2
And
0
4
6
1b
And
0
600
800
1000
200
400
T (°C)
12
East
9
12
Ky
Filchnerfjella and Orvinfjella Ky
10 8
7
8
2
6
6 2
4
Sil
West
2 200
1000
14
8
0
800
P (kbar)
H.U. Sverdrupfjella
10
600 T (°C)
P (kbar) 14
11
3
Sil
2
And
400
4
And
0
600
800
1000
200
400
T (°C) 1a & b: Bucher-Nurminen & Ohta (1993) 2: Grantham et al. (1995)
600
1000
T (°C) 8: Engvik & Elvevold (2004) 9: Board et al. (2005)
3: Rameshwar Rao et al. (1997)
10: Bisnath & Frimmel (2005)
4: Ravikant & Kundu (1998)
11: Baba et al. (2006)
5: Dasgupta et al. (2001)
1.0–1.1 Ga
6: Owada et al. (2003)
0.55–0.50 Ga c. 1.0–1.1 or c. 0.55 Ga
7: Colombo & Taralico (2004)
800
This study
Fig. 8. P– T paths previously reported from CDML.
presence of a decompressional path is obvious. In Orvinfjella, Colombo & Talarico (2004) also described a clockwise P–T path (Fig. 8, path 7) associated with late decompression, and reported a closely similar decompressional texture with orthopyroxene þ plagioclase symplectite after garnet þ hornblende (Colombo & Talarico 2004, plate 3, fig. 2). According to these recent reports, we conclude that a decompressional metamorphic history is characteristic for CDML nunataks.
Grantham et al. (1995) have proposed two stages of metamorphism: an early stage of Grenvillian age (1000–1200 Ma) and a later stage of Pan-African age (500 Ma) in H.U. Sverdrupfjella located east of Jutulsessen. They revealed two different metamorphic paths in the west and east regions, as shown in Figure 8 (path 2). Recently, Board et al. (2005) reported relics of eclogite-facies garnet– omphacite assemblage indicating that peak metamorphic conditions recorded by most rocks in the
CONTRASTING METAMORPHIC P– T PATH IN CDML
415
Fig. 9. Photomicrograph showing symplectitic orthopyroxene in mafic gneiss from Hochlinfjella region. Garnet is not present in this thin section. Scale bar represents 1.0 mm.
area were attained subsequent to decompression from 687–758 8C and .12 kbar (Fig. 8, path 9). They concluded that the high-P metamorphism can be ascribed to subduction and accretion around 565 Ma on the basis of U–Pb single zircon age data. Decompression from the high-P condition is coeval with that observed in Filchnerfjella and Jutulsessen, although the temperature conditions are slightly lower. Formation of hornblende and plagioclase symplectite replacing garnet and clinopyroxene is a characteristic feature of metamorphic rocks in H.U. Sverdrupfjella (Grantham et al. 1995; Board et al. 2005). One possible interpretation for the variation in temperature is the difference in the volume of subsequent granitoid activity in the various areas; however, this needs further assessment. In the Schirmacher Hills, we emphasize that the metamorphic history has an IBC retrograde path after UHT metamorphism (Fig. 8, path 11). This metamorphic process was established in the eastern part of the area, although a metamorphic texture indicating IBC retrograde paths was reported by Dasgupta et al. (2001) (Fig. 8, path 5) and Ravikant & Kundu (1998) (Fig. 8, path 4) in the central part of the area. Therefore, an IBC metamorphic history is a characteristic feature for high-grade rocks in the Schirmacher Hills. A fluid inclusion study by Rameshwar Rao et al. (1997) gave a moderate gradient of the retrograde path (Fig. 8, path 3), but did not give a precise evolution of rock textures, thus we are unable to discuss the results here.
From the viewpoint of metamorphism, it can be concluded that high-grade rocks of Jutulsessen and Filchnerfjella experienced a common metamorphism involving a near-ITD history in their retrograde stage. This ITD history in rocks may be traceable in an east –west direction, and probably includes high-grade rocks of Hochlinfjellet and further west of Orvinfjella and Wohlthatmassiv. In the case of H.U. Sverdrupfjella, slightly high-P and low-T conditions have been reported compared with the eastern region; however, this area shows a decompression history that may imply the westward continuation of this ITD. On the other hand, the rocks from the Schirmacher Hills underwent a different history. At present, it is premature to provide a definite conclusion on correlation between the three regions because there are no reliable age data. The Sm– Nd mineral isochron age from the Schirmacher Hills (Ravikant et al. 2004) shows that the probable retrograde metamorphism is older than the earlier reported 570 Ma granulite-facies metamorphism in CDML (Jacobs et al. 1998). This older metamorphic age compared with the Pan-African metamorphic age recorded in CDML is also evidenced by sensitive high-resolution ion microprobe (SHIRIMP) U –Pb zircon age dating of sapphirine-bearing UHT granulite in the Schirmacher Hills (c. 765 Ma: Baba et al., unpubl. data). We can therefore conclude that the rocks of the Schirmacher Hills had a different evolution
416
S. BABA ET AL.
history and formed independently of those of the Jutulsessen–Filchnerfjella region, as they preserve a different metamorphic history prior to the Gondwana assembly. It is likely that the Schirmacher Hills is an isolated terrane belonging to East Africa (e.g. Lurio belt: Ravikant et al. 2004), where a 700–800 Ma tectonothermal event has been recorded. The metamorphic rocks in the CDML nunataks are interpreted to have formed via collision of the Grenvillian-age crust (or arc) during the Pan-African orogeny. U–Pb zircon age data showing Grenvillian protolith and metamorphic ages from Gjelsvikfjella, Orvinfjella and Wohlthatmassiv may represent the age of preexisting Grenvillian-age crust and metamorphism, in a magmatic arc setting, prior to the collision of East and West Gondwana. This collision is indicated by the preservation of retrograde isothermal decompression metamorphic history, which represents isostatic rebound or rapid extension from an overthickened crust. Such an overthickened crust continues in an east–west direction for at least c. 150 km (Jutulsessen to Filchnerfjella), and is comparable with that of an orogenic mountain chain like the present-day Himalayan collisional orogen. The majority of zircon-rim U–Pb ages in the range of 500–550 Ma from metamorphic rocks in CDML (Jacobs et al. 1998, 2003a) are interpreted to reflect the timing of this metamorphism related to collision and subsequent post-collisional magmatism. In particular, decompression accompanied by slight heating is more likely to be a result of extensional deformation (e.g. the D4 shearing of Bauer et al. 2003) and subsequent syenite emplacement. The present paper is based on fieldwork carried out while at Novolazarevskaya Station and Troll Station during the Japan–Germany–Norway joint expedition in the Mu¨hlig-Hofmannfjella region in 2001–2002. We would like to thank S. Elvevold, A. Lau¨fer and I. Manson for their co-operation in the fieldwork. We acknowledge the logistical support provided by the Norwegian Polar Institute, Alfred-Wegener-Institute for Polar and Marine Research (Germany) and National Institute of Polar Research (Japan). G. H. Grantham, D. Dunkley and V. Ravikant are thanked for their reviews of the manuscript and useful comments. This work was partly supported by a Grant-in-Aid for Scientific Research from Japan Society for Promotion of Sciences to S.B. (No. 16740289).
References A USTRHEIM , H., E LVEVOLD , S., E NGVIK , A. K. & P AULSSON , O. 1997. Geological observations in Gjelsvikfjella, Mu¨hlig-Hofmannfjella and western Orvinfjella during NARE 96/97. Norsk Polarinstitutt Meddelelser, 148, 77–84. B ABA , S., O WADA , M., G REW , E. S. & S HIRAISHI , K. 2006. Sapphirine granulite from Schirmacher Hills, Central Dronning Maud Land. In: F U¨ TTERER , D. K.,
D AMASKE , D., K LEINSCHMIDT , G., M ILLER , H. & T ESSENSOHN , F. (eds) Antarctica: Contributions to Global Earth Science, Springer, Berlin, 37– 44. B AUER , W., T HOMAS , R. J. & J ACOBS , J. 2003. Proterozoic-Cambrian history of Dronning Maud Land in the context of Gondwana assembly. In: Y OSHIDA , M., W INDLEY , B. F. & D ASGUPTA , S. (eds) Proterozoic East Gondwana: Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 247–269. B HATTACHARYA , A., K RISHNAKUMAR , K. R., R AITH , M. & S EN , S. K. 1991. An improved set of a–X parameters for Fe– Mg–Ca garnets and refinements of the orthopyroxene –garnet thermometer and the orthopyroxene–garnet–plagioclase–quartz barometer. Journal of Petrology, 32, 629–656. B ISNATH , A. & F RIMMEL , H. E. 2005. Metamorphic evolution of the Maud Belt: P –T –t path for highgrade gneiss in Gjelsvikfjella, Dronning Maud Land, East Antarctica. Journal of African Earth Sciences, 43, 505–524. B ISNATH , A., F RIMMEL , H. E., A RMSTRONG , R. A. & B OARD , W. S. 2006. Tectono-thermal evolution of the Maud Belt: New SHRIMP U–Pb zircon data from Gjelsvikfjella, Dronning Maud Land, East Antarctica. Precambrian Research, 150, 95–121. B OARD , W. S., F RIMMEL , H. E. & A RMSTRONG , R. A. 2005. Pan-African tectonism in the Western Maud Belt: P –T –t path for high-grade gneisses in the H.U. Sverdrupfjella, East Antarctica. Journal of Petrology, 46, 671–699. B UCHER -N URMINEN , K. & O HTA , Y. 1993. Granulite and garnet–cordierite gneisses from Dronning Maud Land, Antarctica. Journal of Metamorphic Geology, 11, 691–703. C ARSWELL , D. A. & H ARLEY , S. L. 1989. Mineral thermometry and barometry. In: C ARSWELL , D. A. (ed.) Eclogite Facies Rocks. Blackie, Glasgow, 83–110. C OLOMBO , F. & T ALARICO , F. 2004. Regional metamophism in the high-grade basement of central Dronning Maud Land, East Antarctica. Geologisches Jahrbuch, Reihe B, B96, 7– 47. D ASGUPTA , S., S ENGUPTA , S., B OSE , S., F UKUOKA , M. & D ASGUPTA , S. 2001. Polymetamorphism in the Schirmacher Hills granulites, East Antarctica: implications for tectonothermal reworking of an isobarically cooled deep continental crust. Gondwana Research, 4, 337–357. E CKERT , J. O., N EWTON , R. C. & K LEPPA , O. J. 1991. The DH of reaction and recalibration of garnet– pyroxene– plagioclase–quartz geobarometer in the CMAS system by solution calorimetry. American Mineralogist, 76, 148–160. E LLIS , D. J. & G REEN , D. H. 1979. An experimental study of the effect of Ca upon garnet– clinopyroxene Fe–Mg exchange equilibria. Contributions to Mineralogy and Petrology, 71, 13–22. E NGVIK , A. K. & E LVEVOLD , S. 2004. Pan-African extension and near-isothermal exhumation of a granulite facies terrain, Dronning Maud Land, Antarctica. Geological Magazine, 141, 649– 660. F ITZSIMONS , I. C. W. & H ARLEY , S. L. 1994. The influence of retrograde cation exchange on granulite P –T
CONTRASTING METAMORPHIC P– T PATH IN CDML estimates and a convergence technique for the recovery of peak metamorphic conditions. Journal of Petrology, 35, 543– 576. G ANGULY , J., C HENG , W. & T IRONE , M. 1996. Thermodynamics of aluminosilicate garnet solid solution: new experimental data, an optimized model, and thermometric applications. Contributions to Mineralogy and Petrology, 126, 137– 151. G RANTHAM , G. H., J ACKSON , C., M OYES , A. B., G ROENEWALD , P. B., H ARRIS , P. D., F ERRAR , G. & K RYNAUW , J. R. 1995. The tectonothermal evolution of the Kirwanveggen–H.U. Sverdrupfjella areas, Dronning Maud Land, Antarctica. Precambrian Research, 75, 209–229. G REEN , D. H. & R INGWOOD , A. E. 1967. An experimental invistigation of the gabbro to eclogite transformation and its petrological applications. Geochimica et Cosmochimica Acta, 31, 767–833. G REW , E. S. & M ANTON , W. I. 1983. Geochronologic studies in East Antarctica: Reconnaissance uranium/ thorium/lead data from rocks in the Schirmacher Hills and Mount Stinear. Antarctic Journal of the United States, 18, 6 –8. H ARLEY , S. L. 1984. An experimental study of the partitioning of Fe and Mg between garnet and orthopyroxene. Contributions to Mineralogy and Petrology, 86, 359–373. H ARLEY , S. L. 1989. The origin of granulite: a metamorphic perspective. Geological Magazine, 126, 215–247. J ACOBS , J., F ANNING , C. M., H ENJES -K UNST , F., O LESCH , M. & P AECH , H.-J. 1998. Continuation of the Mozambique Belt into East Antarctica: Grenville age metamorphism and polyphase Pan-Africa high grade events in Central Dronning Maud Land. Journal of Geology, 106, 385–406. J ACOBS , J., F ANNING , C. M. & B AUER , W. 2003a. Timing of Grenville-age vs. Pan-African medium- to high-grade metamorphism in western Dronning Maud Land (East Antarctica) and significance for correlations in Rodinia and Gondwana. Precambrian Research, 125, 1 –20. J ACOBS , J., K LEMD , R., F ANNING , C. M., B AUER , W. & C OLOMBO , F. 2003b. Extensional collapse of the late Neoproterozoic–early Palaeozoic East Africa– Antarctic Orogen in central Dronning Maud Land, East Antarctica. In: Y OSHIDA , M., W INDLEY , B. F. & D ASGUPTA , S. (eds) Proterozoic East Gondwana: Supercontinent Assembly and Breakup. Geological Society, London, Special Publications, 206, 271–287. K RETZ , R. 1983. Symbols for rock-forming minerals. American Mineralogist, 68, 277–279. K ROGH , E. J. 1988. The garnet–clinopyroxene Fe– Mg geothermometer—a reinterpretation of existing experimental data. Contributions to Mineralogy and Petrology, 99, 44–48. L EE , H. Y. & G ANGULY , J. 1988. Equilibrium compositions of coexisting garnet and orthopyroxene: experimental determinations in the system FeO–MgO– Al2O3 – SiO2, and applications. Journal of Petrology, 29, 93– 113.
417
M OECHER , D. P., E SSENE , E. J. & A NOVITZ , L. M. 1988. Calculation and application of clinopyroxene – garnet– plagioclase –quartz geobarometers. Contributions to Mineralogy and Petrology, 100, 92– 106. O HTA , Y. 1999. Nature environment map Gjelsvikfjella and western Mu¨hlig-Hofmannfjella, Dronning Maud Land, East Antarctica, 1:100 000. Norsk Polarinstitutt, Temakart, 24. O WADA , M., B ABA , S., L A¨ UFER , A., E LVEBOLD , S., S HIRAISHI , K. & J ACOBS , J. 2003. Geology of eastern Mu¨hlig-Hofmannfjella and Filchnerfjella in Dronning Maud Land, East Antarctica: A preliminary report on a Japan–Norway–Germany joint geological investigation. Polar Geosciences, 16, 108–136. P AULSSON , O. & A USTRHEIM , H. 2003. A geochronological and geochemical study of rocks from Gjelsvikfjella, Dronning Maud Land, Antarctica—implications for Mesoproterozoic correlations and assembly of Gondwana. Precambrian Research, 125, 113–138. P ERKINS , D. & C HIPERA , S. J. 1985. Garnet– orthopyroxene– plagioclase –quartz barometry: refinement and application to the English River subprovince and the Minnesota River valley. Contributions to Mineralogy and Petrology, 89, 69–80. P ERKINS , D. & N EWTON , R. C. 1981. Charnockite geobarometers based on coexisting garnet–pyroxene– plagioclase– quartz. Nature, 292, 144– 146. P IAZOLO , S. & M ARKL , G. 1999. Humite- and scapolitebearing assemblages in marbles and calcsilicates of Dronning Maud Land, Antarctica: new data for Gondwana reconstructions. Journal of Metamorphic Geology, 17, 91– 107. R AMESHWAR R AO , D., S HARMA , R. & G URURAJAN , N. S. 1997. Mafic granulites of Schirmacher region, East Antarctica: fluid inclusion and geothermobarometric studies focusing on the Proterozoic evolution of crust. Transactions of the Royal Society of Edinburgh, Earth Sciences, 88, 1 –17. R AVIKANT , V. 2005. Metamorphism of ultramafic and mafic enclaves within granulites, Schirmacher Oasis, East Antarctica. Journal of the Geological Society of India, 65, 279– 290. R AVIKANT , V. & K UNDU , A. 1998. Reaction textures of retrograde pressure– temperature– deformation paths from granulites of Schirmacher Hills, East Antarctica. Journal of the Geological Society of India, 51, 305– 314. R AVIKANT , V., B HASKAR R AO , Y. J. & G OPALAN , K. 2004. Schirmacher Oasis as extension of the Neoproterozoic East African orogen into Antarctica: New Sm– Nd isochron age constraints. Journal of Geology, 112, 607–616. S ENGUPTA , S. 1993. Tectonothermal history recorded in mafic dykes and enclaves of gneissic basement in the Schirmacher Hills, East Antarctica. Precambrian Research, 63, 273– 291. S TERN , R. J. 1994. Arc assembly and continental collision in the Neoproterozoic East Africa Orogen: Implications for the consolidation of Gondwana. Annual Review of Earth and Planetary Sciences, 22, 319–351.
Empirical thermometer of TiO2 in quartz for ultrahigh-temperature granulites of East Antarctica TOSHISUKE KAWASAKI1 & YASUHITO OSANAI2 1
Department of Earth Sciences, Graduate School of Science and Engineering, Ehime University, Bunkyo-cho 2-5, Matsuyama 790-8577, Japan (e-mail:
[email protected])
2
Division of Evolution of Earth Environments, Graduate School of Social and Cultural Studies, Kyushu University, Ropponmatsu 4-2-1, Fukuoka, 810-8560, Japan Abstract: Two preliminary experiments, heating of rutilated quartz grains with 0.082 wt% TiO2 on average from Bunt Island, Napier Complex, East Antarctica and a synthetic TiO2 – SiO2 (rutile–cristobalite) system in air at 1300 8C for 39 days, showed increasing solubility of TiO2 in silica minerals with temperature. Bunt quartz was converted to cristobalite and traversed by many transparent seams with the disappearence of needles and spots of rutile. Unreacted host grains retained many fine needles of rutile. The seams are homogeneous and slightly enriched in TiO2 up to 0.149 wt% on average, which is about one-fifth lower than that of the synthesized TiO2 –SiO2 cristobalite (0.767 wt% on average). Area analyses with an electron beam in the raster mode at a magnification of 5000 gave 0.308 wt% TiO2 for the bulk composition of the Bunt quartz. This indicates that needles of rutile exsolved from the TiO2-saturated quartz at the cooling stage, or during retrograde metamorphism. Natural examples of quartz in geologically and petrologically well-characterized metamorphic rocks were chemically analysed to examine the temperature controls on the Ti saturation level in quartz. The TiO2 content of quartz in equilibrium with rutile increases sensitively with the metamorphic temperature, which can be expressed as Qtz þ 1:729Þ 273 Tð CÞ ¼ 5895=ðln XTiO 2 W
X Qtz TiO2
is the mole fraction of TiO2, or the number of Ti atoms per formula unit based on where a two-oxygen atom normalization. This empirical equation is very useful to evaluate the metamorphic temperatures for ultrahigh-temperature granulites. The temperatures calculated by the existing Ti-in-quartz thermometer are about 200 8C higher than those estimated by the present thermometer, potentially because of underestimates of Ti solubility in quartz in the previous calibration.
Rutile lamellae are often observed in porphyroblastic quartz, orthopyroxene, garnet, sapphirine, osumilite and mesoperthite in ultrahigh-temperature granulites from the Napier Complex, East Antarctica (Kawasaki & Motoyoshi 2000; Osanai & Yoshimura 2002; Wark & Watson 2006), and are also found in garnet and clinopyroxene in ultrahigh-pressure eclogites and peridotites (Dobrzhinetskaya et al. 1996; Song et al. 2004). In extreme cases of ultrahigh-temperature metamorphism, a considerable amount of TiO2 could dissolve into silicates at the peak metamorphic conditions, and during the subsequent retrograde metamorphism (decompression or cooling) rutile would precipitate from the TiO2-saturated hosts (Kawasaki & Motoyoshi 2007). Recently, Wark & Watson (2006) experimentally calibrated the Ti-in-quartz thermometer (TitaniQ) at 10 kbar and 600 –1000 8C and estimated an equilibration temperature of 992 8C for the quartz from Mt. Riiser-Larsen, Napier Complex, East Antarctica,
which contains exsolution lamellae of rutile formed at the cooling stage of ultrahigh-temperature metamorphism. This is about 200 8C higher than temperatures for retrograde metamorphism of ultrahigh-temperature granulites (e.g. Osanai et al. 2001; Kawasaki et al. 2002; Kawasaki & Motoyoshi 2005; Sato et al. 2008). In this paper, we report the data of an annealing experiment on rutilated quartz from Bunt Island, Napier Complex, East Antarctica, and a synthetic experiment on cristobalite in the TiO2 –SiO2 system, further proving the solubility of Ti in silica minerals. We also present the dataset for natural quartz from geologically and petrologically well-characterized metamorphic rocks including the intermediate-grade and high-pressure metamorphosed quartz eclogite from the Sanbagawa belt, central Shikoku, Japan and ultrahigh-temperature granulites from Napier Complex (Bunt Island, McIntyre Island and Mt. Riiser-Larsen) and Lu¨tzow-Holm Complex (Rundva˚gshetta), East
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 419 –430. DOI: 10.1144/SP308.21 0305-8719/08/$15.00 # The Geological Society of London 2008.
420
T. KAWASAKI & Y. OSANAI
Antarctica. The datasets indicate increasing solubility of TiO2 with temperature. To explain this, we propose a thermodynamic model of rutile – quartz equilibrium. Lastly, we present a new geothermometer based on the TiO2 content in quartz coexisting with rutile for the ultrahightemperature granulites.
As is widely recognized, Ti is in sixfold coordination with oxygen and its electric charge is þ4, whereas Si with þ4 charge occupies a fourfold position in silicate minerals. Titanium could seldom replace the quadrivalent silicon in the SiO4 tetrahedron. Rankama & Sahama (1950) pointed out that a very small amount of Ti could substitute for Si, and occupies the fourfold position in silicates, from geochemical observations of minerals. Recently, many workers (Mosley 1981; Drury & van Roermund 1988; Green et al. 1997; Hermann et al. 2005) have suggested tetrahedral substitution of Ti in olivine. Titanium and silicon are refractory elements to substitute for each other in both quartz and rutile. Although silica minerals show very little atomic substitution, higher temperatures permit a somewhat greater tolerance. We propose a thermodynamic model for TiO2 solubility in quartz coexisting with rutile. To simplify the discussion, a binary TiO2 –SiO2 system is considered. The equilibrium between these phases is described by the following chemical reactions: SiO2 O SiO2
(1)
TiO2 O TiO2
(2)
Rt
and Rt
Qtz
Rt XTiO 1 and 2 Rt ’1 XTiO 2
(3)
where X is the mole fraction, or the number of atoms per formula unit based on a two-oxygen atom normalization. The chemical reactions (1) and (2) indicate the energetic relations among the chemical potentials m of the SiO2 and TiO2 components of quartz and rutile: Rt mQtz SiO2 ¼ mSiO2
Rt and mQtz TiO2 ¼ mTiO2
Rt Rt and aRt SiO2 ¼ kSiO2 XSiO2 (5) (Henry’s law)
Qtz aQtz SiO2 ¼ XSiO2
Rt aRt TiO2 ¼ XTiO2 (Raoult’s law)
and
(6)
where a and k denote the activity and Henry’s coefficient, respectively. The free energy changes of Si and Ti partitioning, DG8SiO2 and DG8TiO2 of reactions (1) and (2), are given by the following equations: ,Qtz ,Rt mSiO ) ¼ RT ln DGSiO2 ¼ (mSiO 2 2 W
W
W
¼ RT ln
Rt kSiO X Rt 2 SiO2 Qtz XSiO 2
aRt SiO2 aQtz SiO2
Rt ’ RT ln kSiO X Rt (7) 2 SiO2
and ,Qtz ,Rt mTiO ) ¼ RT ln DGTiO2 ¼ (mTiO 2 2 W
W
W
¼ RT ln
Rt XTiO 2 Qtz Qtz kTiO2 XTiO 2
aRt TiO2 aQtz TiO2
Qtz ’ RT ln kTiO X Qtz (8) 2 TiO2
where m8, R and T are the standard-state chemical potential, gas constant and absolute temperature, respectively. We note that both rutile and quartz are almost pure phase, Qtz Rt ’ 1 and XTiO ’1 XSiO 2 2
Here the contents of TiO2 in quartz and SiO2 in rutile are very low, so that Qtz Qtz ’ 1, XTiO 1, XSiO 2 2
Qtz Qtz aQtz TiO2 ¼ kTiO2 XTiO2
whereas condensed components of SiO2 in quartz and TiO2 in rutile follow Raoult’s law:
Thermodynamic background
Qtz
It is assumed that diluted components of TiO2 in quartz and SiO2 in rutile would behave in accordance with Henry’s law:
(4)
These relations give the necessary and sufficient condition for rutile–quartz equilibrium.
as given by equations (3). Equations (7) and (8) give a strict solution for the rutile –quartz equilibrium. As is discussed below, the SiO2 content of rutile is about one-third that of TiO2 in a silica mineral and the Si partitioning is rather insensitive to temperature change compared with Ti partitioning. From this practical aspect, we omit discussion of the Si partitioning. Equation (8) is arranged as an Arrhenius-type equation: Qtz ¼ (DHTiO2 þ PDVTiO2 )=RT ln XTiO 2 W
W
Qtz ) (DSTiO2 =R þ ln kTiO 2 W
(9)
where DH8TiO2, DS8TiO2 and DV8TiO2 are the enthalpy, entropy and volume changes for the Ti
QUARTZ THERMOMETER
partitioning of reaction (2). This equation indicates that it is possible to evaluate the thermodynamic parameters, (DH8TiO2 þ PDV8TiO2)/R and (DS8TiO2/R þ lnk Qtz TiO2), by application of leastsquares calculations if we can obtain a sufficient number of high-quality TiO2 data for quartz in equilibrium at given physical conditions. Conversely, if we evaluate these parameters, we can estimate the equilibrium temperatures using the chemical data for quartz coexisting with rutile. Wark & Watson (2006) presented the TitaniQ thermometer, which resembles equation (9): Qtz ¼ 5:69 3765=T log XTi
(10)
Qtz is given in ppm by weight. where XTi However, equation (10) is not at all equivalent to equation (9), which can be derived from the strict treatment by dilute solution thermodynamic principles (Guggenheim 1952). We must note Qtz that there is a non-linear relation between XTi Qtz and X TiO2:
Qtz XTi ¼
Qtz XTiO ATi 2 Qtz Qtz XTiO (ATi þ 2AO ) þ (1 XTiO )(ASi þ 2AO ) 2 2
where A is the atomic weight. In the case of an extremely dilute solution Qtz Qtz (X Qtz TiO2 1 or X TiO2 ! 0), XTi could be approximately proportional to X Qtz TiO2: ATi 106 Qtz X ASi þ 2AO TiO2
1 ATi ASi Qtz X 1þ ASi þ 2AO TiO2 ATi 106 Qtz ATi ASi Qtz ’ XTiO2 1 XTiO2 ASi þ 2AO ASi þ 2AO ’
ATi 106 Qtz Qtz X ¼ 0:797 106 XTiO 2 ASi þ 2AO TiO2
Chemical analyses Analyses were made with an electron microprobe analyser JEOL model JXA-8800 Superprobe at Ehime University. Standards of known composition included synthetic quartz for Si, rutile for Ti, corundum for Al, eskolaite for Cr, hematite for Fe, manganese oxide for Mn, nickel oxide for Ni, periclase for Mg, wollastonite for Ca, natural albite for Na and natural orthoclase for K. The instrumental conditions were: accelerating voltage 15 kV; electron beam current 0.01 mA; beam diameter 1 –2 mm estimated from the size of contamination spots by excitation during analysis. First-order Ka wavelengths were used for all oxides and data reduction used the ZAF correction procedure. Analysis of Ti was made by 30 s measurements and the others were made by 10 s measurements. The background was measured by setting the spectrometer to both sides of the peak on each spot. The detection limit for Ti was about 200 ppm. The analyses of quartz directly in contact with rutile were made at regions about 10 mm or more from the rutile –quartz grain boundary, to avoid the effects of enhancement by secondary X-ray fluorescence.
106
(11)
Qtz ¼ XTi
421
(12)
where we neglect higher-order terms than the Qtz . second one for XTi Equation (12) is a tangent drawn to the curve Qtz given by equation (11) at X Qtz TiO2 ¼ XTi ¼ 0. This tangent is always located above the curve. The Qtz Qtz and X Ti is practically linear relation between X TiO 2 conserved and equations (9) and (10) have the same Qtz , 0.002. In the case of the type of formula if X TiO 2 ultrahigh-temperature granulites, however, the titanium content of quartz, X Qtz TiO2, often exceeds the critical concentration of 0.002 (see Table 1), so that temperatures estimated by TitaniQ should be higher than those obtained by equation (9).
Experiments Experimental procedures Figure 1a shows abundant rutile lamellae exsolved from porphyroblastic quartz of the ultrahightemperature granulites (sample Bunt 01-5A-Ti) from Bunt Island, Napier Complex, East Antarctica. The rutile lamellae are needle-shaped, up to 50 mm long with basal sections 0.5–2 mm in diameter. We carried out the annealing experiment on the rutilated quartz to examine the hypothesis that during ultrahigh-temperature metamorphic conditions quartz is saturated in TiO2, which subsequently exolved rutile during the retrograde cooling process. Quartz grains of about 1–2 mm size were used for the annealing experiment. We also conducted a synthetic experiment using an oxide gel mixture of TiO2 and SiO2 (1 : 1) as starting material. The rutilated quartz grains were put into a platinum tube, which was filled with a fine-powdered TiO2 – SiO2 gel mixture. One edge of the platinum tube was welded by carbon arc and the other was crimped using fine radio-pliers without welding. The platinum tube was hung by a platinum chain in an electric furnace at Ehime University (Kawasaki 2001). The experiment was carried out in air at 1300 8C for 39 days. Temperature was calibrated by the melting point of gold at 1064.18 8C and was measured by the use of a Pt/Pt–13%Rh
422
Table 1. Mean chemical compositions of quartz Locality: Sample no: Occurrence of Qtz:
In contact with Rt* 99.508(795) 0.076(28) 0.015(10) 0.040(28) 0.099(50) 0.027(17) 0.014(8) 0.034(24) 0.024(14) 0.020(12) 0.024(10) 99.881
Rutilated quartz of Bunt Island Bunt01-5A-Ti
Included in Grt†
Area analysis‡
98.792(782) 0.038(14) 0.012(13) 0.016(23) 0.258(135) 0.013(17) 0.010(17) 0.018(26) 0.025(29) 0.013(17) 0.005(8) 99.200
98.256(837) 0.308(56) 0.395(216) 0.022(12) 0.101(101) 0.037(28) 0.160(148) 0.048(29) 0.035(24) 0.061(24) 0.122(71) 99.545
Seam
99.947(720) 0.149(43) 0.163(41) 0.016(25) 0.013(14) 0.015(19) 0.020(16) 0.015(18) 0.008(14) 0.017(14) 0.054(28) 100.417
Host
99.649(735) 0.082(30) 0.020(13) 0.051(39) 0.035(28) 0.029(26) 0.014(9) 0.034(20) 0.035(20) 0.023(16) 0.025(16) 99.997
McIntyre Island SP93022004A In contact with Rt* 99.122(924) 0.314(112) 0.021(13) 0.041(35) 0.186(90) 0.041(25) 0.030(27) 0.037(39) 0.020(12) 0.023(13) 0.016(17) 99.851
Granules within Opx§ 98.920(92) 0.115(32) 0.031(25) 0.035(20) 0.406(155) 0.040(41) 0.028(20) 0.042(21) 0.019(13) 0.020(17) 0.028(17) 99.684
Mt. Riiser Larsen TH96122801 In contact with Rt* 98.602(1245) 0.336(116) 0.025(23) 0.040(34) 0.106(108) 0.030(20) 0.019(14) 0.045(32) 0.023(14) 0.020(13) 0.041(167) 99.287
Included in Opx & Spr} 99.499(912) 0.097(28) 0.027(21) 0.042(35) 0.441(108) 0.028(17) 0.020(28) 0.041(28) 0.029(18) 0.022(14) 0.025(17) 100.271
Rundva˚gshetta RVH92123003A In contact with Rt* 99.572(611) 0.199(43) 0.018(7) 0.014(14) 0.102(68) 0.007(2) 0.004(2) 0.014(6) 0.007(4) 0.005(3) 0.009(4) 99.951
Included in Grt† 99.297(101) 0.109(37) 0.025(6) 0.006(6) 0.223(163) 0.007(4) 0.005(3) 0.013(8) 0.010(4) 0.006(4) 0.010(7) 99.711
Number of cations for 2 oxygens Si Ti Al Cr Fe Mn Mg Ni Ca Na K Total
0.99798(1127) 0.00057(21) 0.00018(13) 0.00032(22) 0.00083(41) 0.00023(14) 0.00021(11) 0.00027(19) 0.00026(16) 0.00039(23) 0.00031(14) 1.00155
0.99800(1118) 0.00029(10) 0.00014(15) 0.00013(18) 0.00218(114) 0.00011(14) 0.00015(27) 0.00015(21) 0.00027(32) 0.00025(35) 0.00006(10) 1.00174
0.99115(1211) 0.00234(42) 0.00470(258) 0.00018(9) 0.00085(86) 0.00032(23) 0.00241(222) 0.00039(24) 0.00038(25) 0.00119(46) 0.00157(92) 1.00546
0.99673(1015) 0.00112(32) 0.00192(48) 0.00013(19) 0.00011(12) 0.00013(16) 0.00030(23) 0.00012(15) 0.00009(14) 0.00033(29) 0.00069(36) 1.00164
0.99801(1040) 0.00062(23) 0.00024(15) 0.00040(31) 0.00029(23) 0.00025(22) 0.00021(14) 0.00027(16) 0.00038(22) 0.00045(31) 0.00032(20) 1.00143
0.99560(1313) 0.00237(85) 0.00025(15) 0.00033(28) 0.00156(75) 0.00035(21) 0.00045(40) 0.00030(31) 0.00022(12) 0.00045(25) 0.00021(21) 1.00207
0.99609(152) 0.00087(24) 0.00037(29) 0.00028(16) 0.00342(130) 0.00034(34) 0.00042(31) 0.00034(17) 0.00020(14) 0.00039(33) 0.00036(21) 1.00309
0.99573(1778) 0.00255(88) 0.00030(27) 0.00032(27) 0.00090(91) 0.00026(17) 0.00029(22) 0.00037(25) 0.00025(15) 0.00039(25) 0.00053(216) 1.00187
0.99616(1291) 0.00073(21) 0.00032(25) 0.00033(28) 0.00369(91) 0.00024(14) 0.00030(43) 0.00033(22) 0.00031(20) 0.00043(29) 0.00032(21) 1.00316
0.99763(852) 0.00150(32) 0.00021(8) 0.00011(10) 0.00086(56) 0.00006(1) 0.00006(2) 0.00011(4) 0.00007(4) 0.00010(5) 0.00011(5) 1.00082
0.99775(158) 0.00083(27) 0.00029(6) 0.00005(4) 0.00188(134) 0.00006(3) 0.00007(4) 0.00010(7) 0.00011(5) 0.00012(7) 0.00013(8) 1.00137
Rutilated quartz grains from Bunt Island were put into a fine-powdered TiO2 –SiO2 gel mixture in a platinum capsule, and heated at atmospheric pressure and 1300 8C for 936 h (run 010619). *Quartz was in direct contact with rutile. †Quartz was included within garnet porphyroblasts and contained exsolution lamellae of rutile. ‡Area analyses were made with the beam in raster mode at magnification of 5000 on both host and seam. Rutiles were sometimes included in the analysed areas. §Analyses were made on the lamellae and granules of quartz within the orthopyroxene porphyroblasts. Granules of quartz contained the exsolution lamellae of rutile. }Analyses were made on the quartz inclusions within orthopyroxene porphyroblasts and/or sapphirine –quartz intergrowth.
T. KAWASAKI & Y. OSANAI
SiO2 TiO2 Al2O3 Cr2O3 FeO MnO MgO NiO CaO Na2O K2 O Total
Quartz eclogite mass 99050804QE
QUARTZ THERMOMETER
423
Fig. 1. Photomicrographs showing quartz from Bunt Island, Napier Complex, East Antarctica and run products of run 010619 at 1 atm and 1300 8C for 936 h. (a) Rutilated quartz in plane-polarized, transmitted light. Many needle rutiles, length of which is about 50 mm or more, are observed in quartz. (b) Back-scattered electron (BSE) image of the run product annealed in the air at 1300 8C for 936 h. Thin white seams divide the grain. (c) BSE image of the run product (enlarged). Fine rutile needles and spots are included within host. The white seam including rare rutile is homogeneous compared with the host. (d) BSE image of cristobalite and rutile in the TiO2 –SiO2 system synthesized at 1 atm and 1300 8C for 936 h.
thermocouple placed at the top of the sample space and at the outer side in contact with the reaction tube in the furnace. The run temperature was controlled by the measurement of the outer-side temperature of the reaction tube, and was kept constant within +1 8C of the nominal values. After the heat treatment, samples were quenched by dropping into a beaker fixed with distilled water at the bottom of the reaction tube. The run products in the platinum container were mounted in epoxy resin and polished for examination with the electron microprobe.
Run products The run product of the annealing experiment (run 010619A) is shown, as backscattered electron (BSE) images of the annealed grains, in Figure 1b and c. Synthetic cristobalite and rutile are shown in Figure 1d. Many white seams divide the dark
host. We identified both seam and host as cristobalite by micro-focused X-ray diffraction experiments by means of an MAC Science M21X1 instrument at Ehime University, although at atmospheric pressure and 1300 8C tridymite is the stable silica mineral based on thermodynamic calculation (Swamy et al. 1994). As is observed in a phase diagram of the AlPO4 –SiO2 system (Veksler et al. 2003), a dilute component in SiO2 would reduce the tridymite –cristobalite transition temperature. It was confirmed that the white seam is the reacted part and the dark host is the unreacted part, which retains numerous needles and spots of rutile. Although some rutile is found within the white seam, most of the fine needles and spots of rutile had vanished from the seams. The white seam is rather homogeneous when compared with the dark host (see Fig. 1b and c). Chemical compositions of the host and the seam are given in Table 1 and Figures 2 and 3. The mean Ti content of the
424
T. KAWASAKI & Y. OSANAI
Fig. 2. Frequency histogram of number of Ti atoms per formula unit (O ¼ 2) in quartz and cristobalite. Each block equals one analysis. Mean Ti contents are shown by arrows. Standard errors are given in parentheses and refer to the last decimal place. The 0.001 Ti atoms line is dotted to emphasize separation between high-Ti quartz of the ultrahightemperature granulites at the peak and low-Ti quartz from the quartz eclogite and the retrograde ultrahightemperature granulites.
QUARTZ THERMOMETER
425
Fig. 3. Frequency histogram of cristobalite Ti and rutile Si contents per formula unit (O ¼ 2) for synthetic run products at an atmospheric pressure and 1300 8C for 936 h. Triangles show the directions in which the contents of Ti in cristobalite and Si in rutile should increase at the final stage. Final compositions of cristobalite and rutile are the Ti-rich and the Si-rich data, respectively. Arrows show mean Ti and Si contents. Standard errors are given in parentheses and refer to the last decimal place.
seam is 0.00112(32) per formula unit. The seam is thus enriched in Ti compared with the unreacted host, which contains 0.00062(23). The mean Ti content of the seam is about one-fifth lower than that of the synthetic cristobalite of 0.0058(18). This indicates that the seam recrystallized in the annealing experiment did not attain equilibrium with rutile, or the bulk TiO2 is lower than the Ti saturation level. The synthetic cristobalite data plot above the Ti-solubility curve of quartz in Figure 4. Cristobalite has a comparatively open
structure, whereas the atoms in quartz are more closely packed. The Ti atoms prefer cristobalite rather than quartz. This is reflected by the higher solubility of Ti in cristobalite than in quartz. We made area analyses of a region about 500 mm 400 mm at 400 different locations including white and dark areas with the electron beam in raster mode at a magnification of 5000. The bulk composition of the rutilated quartz is estimated by the mean of the area analyses as 0.00234 (42) Ti atoms per formula unit (Table 1
Fig. 4. The relation between the Ti content of quartz and the reciprocal of temperature. As is clear, the logarithm of TiO2 in quartz relates to the reciprocal of the temperature. The line shows the result of a least-squares fit on the data: ¼ 25895/T(K) 2 1.729; where cristobalite data were not used for the calculation. W, peak data; * retrograde ln X Qtz TiO2 data; S , cristobalite synthesized at 1 atm and 1300 8C for 936 h. Horizontal and vertical bars show uncertainties of temperature and 1s of Ti contents.
426
T. KAWASAKI & Y. OSANAI
and Fig. 2). According to Osanai et al. (2001), the ultrahigh-temperature granulites of Bunt Island experienced peak metamorphism at temperatures of 1050–1100 8C and pressures of about 9– 10 kbar, and were isothermally decompressed in the sapphirine –quartz stability field, then underwent retrograde metamorphism by isobaric cooling in the garnet–cordierite stability field at about 800 –850 8C. From the present experiments, we confirm that the TiO2 component dissolved into quartz during the progressive metamorphism or at the peak metamorphism, and rutile subsequently crystallized from the TiO2-saturated quartz at the cooling stage, or during the retrograde metamorphism. The bulk composition of rutilated quartz retains the record of the peak metamorphism, and the chemical composition of the dark parts of quartz indicates the retrograde composition of the metamorphism. To examine the property of increased atomic substitution at higher temperatures, we present the TiO2 data for natural quartz in the intermediate-grade and high-pressure metamorphosed quartz eclogite of the Sanbagawa belt, Shikoku, Japan, and ultrahigh-temperature granulites, including the orthopyroxene granulite from McIntyre Island, the sapphirine – quartz granulite from Mt. Riiser-Larsen, Napier Complex, and the quartz – feldspathic granulite from Rundva˚gshetta, Lu¨tzow- Holm Complex, East Antarctica.
ilmenite. Rutile needles crystallized from Ti-saturated quartz within garnet, probably during the retrograde metamorphic stage. The chemical data for quartz are given in Table 1 and Figure 2. The quartz in contact with rutile in the matrix shows distinctly high TiO2 contents compared with that included within garnet, although quartz within garnet is poor in TiO2 because of the precipitation of rutile. The Ti contents of such quartz are less than half those of quartz in the matrix. This indicates that the composition at the peak of the eclogite-facies metamorphism was preserved in the quartz in contact with rutile, and the record of the retrograde stage was imprinted in the composition of quartz with exsolution lamellae of rutile trapped within garnet. These retrograde quartz are distinctly iron-rich (Table 1). The iron atoms in quartz, probably in ferric state, would have an important role in the removal of rutile from Ti-saturated quartz. Kawasaki (1987) estimated a temperature range of 607–660 8C using the olivine –orthopyroxene– garnet geothermometer for the Higashi –Akaishi peridotite mass neighbouring the eclogite mass. This corresponds to epidote– amphibolite-facies metamorphism (Takasu 1989). We assume that this is the condition of the retrograde metamorphism. The TitaniQ thermometer (Wark & Watson 2006) gives temperatures of 920 and 818 8C for the peak and retrograde stages, respectively (Table 2). These temperatures are much too high for the Sambagawa metamorphism.
TiO2 contents of natural quartz Quartz eclogite from Sanbagawa metamorphic belt, central Shikoku, Japan We present the chemical data for quartz coexisting with rutile in the quartz eclogite (sample 99050804QE) from the Sanbagawa metamorphic belt, central Shikoku, Japan as an example in the intermediate temperature range. The quartz eclogite mass is located between the Jiyoshi-yama peridotite mass and the Higashi-Akaishi peridotite mass (Kugimiya & Takasu 2002). According to Takasu (1989), the source rock of the quartz eclogite is sedimentary in origin, composed of alternations of pelitic and basaltic volcanoclastic rocks. These rocks underwent progressive metamorphism to eclogite facies, the peak condition of which was estimated as 700 –750 8C at 17 kbar. After peak metamorphism, this eclogite underwent retrograde metamorphism in the epidote– amphibolite facies. The eclogite is composed of garnet, omphacite, quartz, kyanite, phengite, epidote and rutile. The quartz occur in contact with rutile and/or as inclusions within garnet. Garnet includes quartz, which occasionally contains needles of rutile and
Orthopyroxene granulite from McIntyre Island, Enderby Land, East Antarctica The McIntyre granulite (sample SP93022004A) consists of orthopyroxene, garnet and quartz, with minor sillimanite and accessory amounts of rutile, phlogopite and zircon. Garnet occurrs as small euhedral blebs, exsolution lamellae in orthopyroxene and thin films wrapping orthopyroxene, euhedral sillimanite, rutile and phlogopite (Kawasaki & Motoyoshi 2000). Quartz is enclosed in orthopyroxene porphyroblasts, and is elongated in the same direction as the garnet lamella, and/or occurs as rounded blebs including rutile needles. In very rare cases, quartz is trapped in orthopyroxene, which is in direct contact with rutile. Rutile occurs as needle-shaped and granular crystals. Granular rutile is moated by garnet or is in contact with garnet blebs and quartz in orthopyroxene. Rutile needle are not surrounded by garnet, and mainly occur in orthopyroxene porphyroblasts with the same orientation as the garnet lamellae. This granulite experienced peak metamorphism at temperatures around 1000– 1050 8C (Kawasaki et al. 2002) and was subsequently subjected to
Table 2. Compilation of TiO2 data for quartz, and estimated and calculated temperatures Locality
Rundva˚gshetta Bunt Island McIntyre Island Mt. Riiser-Larsen
Ti (p.f.u. O ¼ 2)
Estimated T(8C)*
Calculated T (8C)†
Calculated T (8C)‡
In contact with Rt Inclusion in Grt} In contact with Rt Inclusion in Grt} Bulk Rutilated Qtz In contact with Rt Granules} In contact with Rt Inclusion in Opx and Spr}
0.00057(21) 0.00029(10) 0.00150(32) 0.00083(27) 0.00234(42)§ 0.00062(23) 0.00237(85) 0.00087(24) 0.00255(88) 0.00073(21)
700 – 750 607 – 660 925 – 1039 778 – 886 1050 – 1100 800 – 850 1000 – 1050 873 – 895 1120 777
754 + 66 646 + 49 962 + 55 826 + 67 1088 + 56 769 + 68 1093 + 113 835 + 57 1117 + 113 800 + 56
920 + 121 818 + 47 1103 + 47 985 + 59 1208 + 45 934 + 62 1211 + 91 994 + 51 1230 + 90 985 + 59
Reference 1 2 3 4 5 5 6 7 8,9 9
QUARTZ THERMOMETER
Mt. Higashi-Akaishi
Occurrence of Qtz
*Temperatures at the peak and retrograde stages were estimated in the literature. 1, Takasu (1989); 2, Kawasaki (1987); 3, Kawasaki & Motoyoshi (2005); 4, Kawasaki & Motoyoshi (2007); 5, Osanai et al. (2001); 6, Kawasaki et al. (2002); 7, Kawasaki & Motoyoshi (2000); 8, Harley & Motoyoshi (2000); 9, Kawasaki & Sato (2002). Qtz Qtz þ 1.729) 2 273, where TiO2 content of quartz, X TiO , is given by mole fraction. †Temperatures were calculated by the following quartz thermometer: T(oC) ¼ 25895/(ln X TiO 2 2 Qtz ‡Temperatures were calculated using the Wark & Watson (2006) thermometer: T(8C) ¼ 23765/(log X Qtz Ti 2 5.69) 2 273, where Ti content of quartz, X Ti , is in ppm by weight. §Area analyses of rectangular region of 500 mm 400 mm were made with the electron beam in raster mode at a magnification of 5000. }Quartz, included within garnet, orthopyroxene or sapphirine, contains rutile needles.
427
428
T. KAWASAKI & Y. OSANAI
retrograde metamorphism at 873 –895 8C, resulting in the formation of garnet lamellae within orthopyroxene and/or garnet moats around orthopyroxene (Kawasaki & Motoyoshi 2000). As is seen in Figure 2, a frequency histogram of the TiO2 content in quartz clearly shows two types: high-Ti quartz that is in direct contact with rutile, and low-Ti quartz that occurs as lamellae and granules sometimes including rutile needles in orthopyroxene. This indicates that high-Ti quartz was in equilibrium with rutile at the stage of peak metamorphism and low-Ti quartz recrystallized and/or unmixed fine rutile during the retrograde stage.
Sapphirine – quartz granulite from Mt. Riiser-Larsen, Enderby Land, East Antarctica Sapphirine–quartz granulite of Mt. Riiser-Larsen (sample TH96122801) shows the highest content of TiO2 in quartz of 0.336 wt% on average. These data were obtained from quartz grains in contact with rutile. The temperature of this granulite is estimated by the orthopyroxene–sapphirine thermometer (Kawasaki & Sato 2002) as 1120 8C, applying to the core Opx composition (XSpr Mg ¼ 0.8392 and XMg ¼ 0.8108), Spr and 777 8C for the rim (XMg ¼ 0.8657 and Opx ¼ 0.7966). The alumina contents of orthopyroxXMg ene are 11.23 wt% at the core and 8.79 wt% at the rim. Orthopyroxene is enriched in MgO and Al2O3 at the core and is slightly Fe-rich and Al2O3-poor at the rim. Metamorphic temperatures exceeding 1100 8C were estimated using the Al2O3 isopleth for the core component of Riiser-Larsen orthopyroxene (Harley & Motoyoshi 2000), which is consistent with that obtained by the orthopyroxene–sapphirine thermometer. Figure 2 indicates that the TiO2 content of quartz directly in contact with rutile is clearly higher than that of inclusions within orthopyroxene and/or sapphirine. These quartz inclusions also contain rutile lamellae. We propose that the high-Ti and low-Ti data give us information on the peak and retrograde metamorphisms, respectively. Wark & Watson (2006) estimated the metamorphic temperature of Mt. Riiser-Larsen granulite as 992 8C for the retrograde quartz containing rutile lamellae. As given in Table 2, the temperature estimates of Wark & Watson’s TitaniQ thermometer are high: 1230 8C and 985 8C for high-Ti and low-Ti quartz, respectively.
Quartzo-feldspathic granulite from Rundva˚gshetta, Lu¨tzow-Holm Complex, East Antarctica This granulite (sample RVH02123003A) was collected during the summer operation (2002–2003)
of the 44th Japanese Antarctic Research Expedition (JARE-44) from Rundva˚gshetta, Lu¨tzow-Holm Bay, East Antarctica. The granulite is a leucocratic and coarse-grained rock, and contains a concordant thin layer of melanocratic sillimanite –cordierite – sapphirine granulite (Kawasaki et al. 1993; Motoyoshi & Ishikawa 1997; Kawasaki & Motoyoshi 2005). The leucocratic granulite consists Grt ¼ 0.341), sillimanite, K-feldspar of garnet (XMg and quartz porphyroblasts, with accessory apatite, ilmenite, monazite, plagioclase, rutile and zircon. Garnet contains many inclusions of quartz, Os ¼ 0.827), and the assemblage spinel osumilite (XMg (Mg/(Mg þ Fe) ¼ 0.537; Zn/(Mg þ Fe þ Zn) ¼ 0.39) þ quartz þ biotite (XBt Mg ¼ 0.767) þ plagioclase. Rutile needles were precipitated in quartz included within garnet porphyroblasts, and are also found in K-feldspar and garnet. Euhedral rectangular rutile is found at grain boundaries of porphyroblasts of quartz, garnet and K-feldspar, and is also found within these minerals. Chemical data for quartz are given in Table 1 and Figure 2. Quartz included in garnet shows systematically low Ti content compared with the porphyroblastic quartz, which is in contact with rutile. The existence of osumilite and the spinel –quartz assemblage as well as a sapphirine –quartz association (Yoshimura et al. 2008) indicates that the Rundva˚gshetta granulite experienced ultrahigh-temperature metamorphism (Kawasaki et al. 1993; Motoyoshi & Ishikawa 1997; Kawasaki & Motoyoshi 2005). Das et al. (2001) reported that osumilite coexisting with spinel is stable at temperatures from 850 to 1000 8C at 7–8.5 kbar. Sajeev & Osanai (2004) reported the occurrence of osumilite and spinel þ quartz inclusions in garnet porphyroblasts of a Sri Lanka granulite and estimated that the pressure – temperature condition must be above 950 8C around 7 kbar from experimental data on the stability field of zincian spinel þ quartz without cordierite (Nichols et al. 1992). We estimated the peak temperature at Rundva˚gshetta as 925– 1039 8C and 11.5–15 kbar (Kawasaki & Motoyoshi 2005). The subsequent retrograde metamorphism with near-isothermal decompression (Motoyoshi & Ishikawa 1997) took place at around 800 8C and 6.5–10.8 kbar (Kawasaki & Motoyoshi 2005). The Ti partitioning in garnet coexisting with rutile and quartz (Kawasaki & Motoyoshi 2007) indicates that the retrograde metamorphic temperatures were 778–886 8C.
Discussion and conclusions The Ti data and the metamorphic temperatures are compiled in Table 2. Figure 4 is an Arrhenius plot of ln X Qtz TiO2 against 1/T. We neglect the pressure
QUARTZ THERMOMETER
effect on TiO2 solubility in quartz because the variation of the metamorphic pressures is limited within a narrow range and the difference in pressure is around 9 kbar at most: 16–25 kbar for the quartz eclogite from the Sanbagawa metamorphic terrane in Japan, and about 10– 15 kbar for the ultrahightemperature granulite from the Napier Complex, East Antarctica. The results of a least-squares fit of the present data in Table 2 are given as (DHTiO2 þ PDVTiO2 )=R ¼ 5895 + 1 and W
W
DSTiO2 =R þ ln kTiO2 ¼ 1:729 + 0:001: W
429
run products. We often discussed the high-temperature metamorphism and the anatexis of lower crustal materials with Y. Yoshimura, M. Owada, Y. Motoyoshi, M. Arima, Y. Hiroi and K. Shiraishi. We obtained the idea of the present study through discussions with them. We express our gratitude to K. Suzuki, M. Obata and A. Takasu for their criticism at an early stage of this study. The manuscript has received constructive criticism from A. Miyake, K. K. Podlesskii and M. Satish-Kumar. Support from the Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture of the Japanese Government No. 14654093 is gratefully acknowledged.
(13)
In this thermodynamic formulation we have omitted the cristobalite data at 1 atm and 1300 8C. Figure 4 shows the results of the least-squares calculation. As can be seen in this figure, the least squares fit is satisfactory. Good fit implies that the TiO2 solubility in quartz coexisting with rutile can be approximated with reasonable precision by an Arrhenius-type equation (9). The assumptions of Raoult’s law for the solvent components SiO2 in quartz and TiO2 in rutile, and of Henry’s law for the solute components such as TiO2 in quartz and SiO2 in rutile are satisfactory. We can easily estimate the metamorphic temperature from TiO2 content in quartz by the empirical equation Qtz T(8C) ¼ 5895=( ln XTiO þ 1:729) 273 (14) 2
This equation indicates that if we can obtain the TiO2 content of quartz coexisting with rutile, we can estimate the temperature for each quartz grain situated in a specific microstructural setting. An approach of this kind has been carried out by Watson & Harrison (2005) with fair success in estimating the crystallization temperature for Hadean zircons with an age of 4.5–4.0 Ga. In Table 2, we also compare the temperatures estimated in the present study with those of Wark & Watson’s (2006) TitaniQ thermometer. The TitaniQ thermometer estimates are about 200 8C higher than those by our calibration. Given the petrological and microstructural evidence in the rocks in this study, we are confident that our empirical calibration provides reliable temperature estimation for different stages in the high-temperature metamorphic evolution of crustal rocks. We express our sincere thanks to members of the 44th Japanese Antarctic Research Expedition for their logistic support and co-operation during the 2002–2003 field season. We thank T. Hokada for his helpful comments and are particularly grateful for generous donation of a valuable sample of Mt. Riiser-Larsen granulite (TH96122801). In the last stage of this study, T. Sanehira kindly helped us in X-ray experiments on
References D AS , K., D ASGUPTA , S. & M IURA , H. 2001. Stability of osumilite coexisting with spinel solid solution in metapelitic granulites at high oxygen fugacity. American Mineralogist, 86, 1423– 1434. D OBRZHINETSKAYA , L. F., G REEN , H. W., II & W ANG , S. 1996. Alpe Arami: a peridotite massif from depths of more than 300 kilometers. Science, 271, 1841–1845. D RURY , M. R. & VAN R OERMUND , H. L. M. 1988. Metasomatic origin for Fe– Ti-rich multiphase inclusions in olivine from kimberlite xenoliths. Geology, 16, 1035– 1038. G REEN , H. W., II, D OBRZHINETSKAYA , L. F., R IGGS , E. M. & J IN , Z.-M. 1997. Alpe Arami: a peridotite massif from the mantle transition zone? Tectonophysics, 279, 1– 21. G UGGENHEIM , E. A. 1952. Mixtures. Oxford University Press, Oxford. H ARLEY , S. L. & M OTOYOSHI , Y. 2000. Al zoning in orthopyroxene in a sapphirine quartzite: evidence for .1120 8C UHT metamorphism in the Napier Complex, Antarctica, and implications for the entropy of sapphirine. Contributions to Mineralogy and Petrology, 138, 293– 307. H ERMANN , J., O’N EILL , H. S. C. & B ERRY , A. J. 2005. Titanium solubility in olivine in the system TiO2 – MgO– SiO2: no evidence for an ultra-deep origin of Ti-bearing olivine. Contributions to Mineralogy and Petrology, 148, 746–760. K AWASAKI , T. 1987. Palaeogeotherms: Olivine– orthopyroxene–garnet geothermometry and geobarometry. Lithos, 20, 263– 278. K AWASAKI , T. 2001. Experimental constraints on mixing properties of forsterite–monticellite olivine at atmospheric pressure. Journal of Mineralogical and Petrological Sciences, 96, 54– 66. K AWASAKI , T. & M OTOYOSHI , Y. 2000. High-pressure and high-temperature phase relations of an orthopyroxene granulite from McIntyre Island, Enderby Land, East Antarctica. Polar Geoscience, 13, 114– 133. K AWASAKI , T. & M OTOYOSHI , Y. 2005. Experimental constraints on the decompressional P –T path of Rundva˚gshetta granulites, Lu¨tzow-Holm Complex, East Antarctica. In: F U¨ TTERER , D. K., D AMASKE , D., K LEINSCHMIDT , G., M ILLER , H. & T ESSENSOHN , F. (eds) Antarctica: Contributions to Global Earth Sciences. Springer, Berlin, 23–36.
430
T. KAWASAKI & Y. OSANAI
KAWASAKI, T. & M OTOYOSHI , Y. 2007. Solubility of TiO2 in garnet and orthopyroxene: Ti thermometer for ultrahightemperature granulites. In: C OOPER , A. & R AYMOND , C. (eds) Antarctica: A Keystone in a Changing World. US Geological Survey Open-File Report, 2007-1047; Short Research Paper, 038, doi:10.3133/of2007-1047.srp038. K AWASAKI , T. & S ATO , K. 2002. Experimental study of Fe–Mg exchange reaction between orthopyroxene and sapphirine and its calibration as a geothermometer. Gondwana Research, 4, 741– 747. K AWASAKI , T., I SHIKAWA , M. & M OTOYOSHI , Y. 1993. A preliminary report on cordierite-bearing assemblages from Rundva˚gshetta, Lu¨tzow-Holm Bay, East Antarctica: Evidence for a decompressional P– T path? Proceedings of NIPR Symposium of Antarctic Geosciences, 6, 47– 56. K AWASAKI , T., S ATO , K. & M OTOYOSHI , Y. 2002. Experimental constraints on the thermal peak of a granulite from McIntyre Island, Enderby Land, East Antarctica. Gondwana Research, 4, 749–756. K UGIMIYA , Y. & T AKASU , A. 2002. Geology of the Western Iratsu mass within the tectonic me´lange zone in the Sanbagawa metamorphic belt, Besshi district, central Shikoku, Japan. Journal of Geological Society of Japan, 108, 644–662 [in Japanese with English abstract]. M OSLEY , D. 1981. Ilmenite exsolution in olivine. American Mineralogist, 66, 976– 979. M OTOYOSHI , Y. & I SHIKAWA , M. 1997. Metamorphic and structural evolution of granulites from Rundva˚gshetta, Lu¨tzow-Holm Bay, East Antarctica. In: R ICCI , C. A. (ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antartic, Siena, 65–72. N ICHOLS , G. T., B ERRY , R. F. & G REEN , D. H. 1992. Internally consistent gahnitic spinel– cordierite– garnet equilibria in the FMASHZn system: geothermobarometry and applications. Contributions to Mineralogy and Petrology, 111, 362 –377. O SANAI , Y. & Y OSHIMURA , Y. 2002. High-temperature limit of crustal metamorphism: a perspective of ultrahightemperature metamorphism. Chishitsu News, 573, 10–26 [in Japanese]. O SANAI , Y., T OYOSHIMA , T., O WADA , M., T SUNOGAE , T., H OKADA , T., C ROWE , W. A. & K USACHI , I. 2001. Ultrahigh temperature sapphirine–osumilite and sapphirine–quartz granulites from Bunt Island in the Napier Complex, East Antarctica—Reconnaissance estimation of P–T evolution. Polar Geoscience, 14, 1–24.
R ANKAMA , K. & S AHAMA , Th. G. 1950. Geochemistry. University of Chicago Press, Chicago, IL. S AJEEV , K. & O SANAI , Y. 2004. Osumilite and spinel þ quartz from Sri Lanka: implications for UHT conditions and retrograde P– T path. Journal of Mineralogical and Petrological Sciences, 99, 320– 327. S ATO , M., M IYAMOTO , T. & K AWASAKI , T. 2008. Fe2þ –Mg partitioning experiments between orthopyroxene and spinel using ultrahigh-temperature granulite from the Napier complex, East Antarctica. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: a Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 431–447. S ONG , S., Z HANG , L. & N IU , Y. 2004. Ultra-deep origin of garnet peridotite from the North Qaidam ultrahighpressure belt, Northern Tibetan Plateau, NW China. American Mineralogist, 89, 1330– 1336. S WAMY , V., S AXENA , S. K., S UNDMAN , B. & Z HANG , J. 1994. A thermodynamic assessment of silica phase diagram. Journal of Geophysical Research, 99, 11787–11794. T AKASU , A. 1989. P– T histories of peridotite and amphibolite tectonic blocks in the Sanbagawa metamorphic belt, Japan. In: D ALY , J. S., C LIFF , R. A. & Y ARDLEY , B. W. D. (eds) Evolution of Metamorphic Belts. Geological Society, London, Special Publications, 43, 533 –538. V EKSLER , I. V., T HOMAS , R. & W IRTH , R. 2003. Crystallization of AlPO4 –SiO2 solid solutions from granitic melt and implications for P-rich melt inclusions in pegmatitic quartz. American Mineralogist, 88, 1724–1730. W ARK , D. A. & W ATSON , E. B. 2006. TitaniQ: a titanium-in-quartz geothermometer. Contributions to Mineralogy and Petrology, 152, 743– 754. W ATSON , E. B. & H ARRISON , T. M. 2005. Zircon thermometer reveals minimum melting conditions on earliest Earth. Science, 308, 841–844. Y OSHIMURA , Y., M OTOYOSHI , Y. & M IYAMOTO , T. 2008. Sapphirine þ quartz association in garnet: implication for ultrahigh-temperature metamorphism in Rundva˚gshetta, Lu¨tzow-Holm Complex, East Antarctica. In: S ATISH -K UMAR , M., M OTOYOSHI , Y., O SANAI , Y., H IROI , Y. & S HIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: a Key to the East–West Gondwana Connection. Geological Society, London, Special Publications, 308, 377– 390.
Fe21 –Mg partitioning experiments between orthopyroxene and spinel using ultrahigh-temperature granulite from the Napier Complex, East Antarctica KEI SATO1, TOMOHARU MIYAMOTO2 & TOSHISUKE KAWASAKI3 1
Research Center for the Evolving Earth and Planets, Department of Earth and Planetary Sciences, Graduate School of Science and Engineering, Tokyo Institute of Technology, Ookayama 2-12-1, Meguro-ku Tokyo 152-8551, Japan (e-mail:
[email protected]) 2
Department of Earth and Planetary Sciences, Faculty of Science, Kyushu University, Hakozaki 6-10-1, Fukuoka 812-8581, Japan 3
Department of Earth Sciences, Graduate School of Science and Engineering, Ehime University, Bunkyo-cho 2-5, Matsuyama 790-8577, Japan
Abstract: Temperature dependence of the Fe2þ – Mg exchange between orthopyroxene (Opx) and spinel (Spl), 1=2Fe2 Si2 O6 þ MgAl2 O4 ¼ 1=2Mg2 Si2 O6 þ FeAl2 O4 Opx
Spl
Opx
Spl
was experimentally determined at 9– 13 kbar and 900–1200 8C using an ultrahigh-temperature (UHT) granulite collected from the Napier Complex in Enderby Land, East Antarctica. The Fe2þ –Mg distribution coefficient, KD ¼
Spl Opx XFe XMg Spl Opx XMg XFe
is empirically obtained as ln KD ¼ 1:134 þ [2341 þ 7:5(P 11)]=T where X is the cationic mole fraction, and pressure P and temperature T are in kbar and Kelvin, respectively. The new geothermometer was applied to various natural UHT and associated high-grade metamorphic rocks from East Antarctica and other regions of the world. The results indicate temperatures between 735 and 902 8C at pressures in the range of 5 –14 kbar. This geothermometer utilizing spinel does not give peak metamorphic condition, because it is relatively easy for spinel to re-equilibrate during the cooling stage of metamorphism. Hence, we conclude that the geothermometer is suitable for evaluating the closure temperature for the KD between aluminous orthopyroxene and spinel during retrograde metamorphism rather than the thermal peak.
Mafic and/or aluminous constituents of ultrahightemperature (UHT) metamorphic rocks are characterized by complex mineral assemblages involving sapphirine, garnet, aluminous orthopyroxene, cordierite, spinel, osumilite and/or F-bearing phlogopite (e.g. Dallwitz 1968; Ellis et al. 1980; Grew 1980; Harley 1998a; Motoyoshi & Hensen 2001). As Harley (1998b) has mentioned, the orthopyroxene– garnet exchange thermometer (Harley & Green 1982; Kawasaki & Matsui 1983; Harley 1984; Lee & Ganguly 1988; Ganguly et al. 1996; Kawasaki & Motoyoshi 2000; Sato & Kawasaki 2002)
is useful for estimation of the temperature of highgrade metamorphism. However, there is a general lack of a geothermometer for the garnet-free system. Kawasaki & Sato (2002) developed an exchange thermometer utilizing the mineral pair of sapphirine and orthopyroxene. Recently, Sato et al. (2006) reported an exchange thermometer using coexisting sapphirine and spinel, based on the experimental data reported by Sato et al. (2004). Liermann & Ganguly (2003) developed an exchange thermometer using the mineral pair of orthopyroxene and spinel. In their partitioning
From: SATISH -K UMAR , M., MOTOYOSHI , Y., OSANAI , Y., HIROI , Y. & SHIRAISHI , K. (eds) Geodynamic Evolution of East Antarctica: A Key to the East– West Gondwana Connection. Geological Society, London, Special Publications, 308, 431 –447. DOI: 10.1144/SP308.22 0305-8719/08/$15.00 # The Geological Society of London 2008.
432
K. SATO ET AL.
experiments, which were performed at 9–14 kbar and 850 –1250 8C, various crystalline mixtures (CM) of synthetic and/or natural Al2O3-poor ¼ 0.08 –0.89 wt%) þ synorthopyroxene (Al2OOpx 3 thetic spinel were used as starting materials. They prepared mixtures of the CM and 5–10 wt% PbO– PbF2 flux for the experiments. The XMg [¼Mg/(Fe2þ þ Mg)] values of orthopyroxene and spinel that constitute the CM are as follows: Spl XOpx Mg ¼ 0.21, 0.49, 0.87 and 1.00; XMg ¼ 0.00, 0.49, 0.51 and 1.00. Natural orthopyroxene crystals often contain Al2O3; in general, a considerable amount of Al2O3 is detected in orthopyroxene within UHT metamorphic rocks (i.e. up to about 13 wt% Al2O3; Harley & Motoyoshi 2000). Thus, Liermann & Ganguly (2003) proposed a correction for the effect of Al in orthopyroxene thermodynamically. However, in their experiments, aluminous orthopyroxenes containing 2–13 mol% alumina were found in the run products. Thus, it is required that olivine is produced by the following Al2O3 buffering reaction, based on the interaction parameters of FeSiO3 –AlAlO3 and MgSiO3 – AlAlO3 (Berman & Aranovich 1996; Kosyakova et al. 2004, 2005): Mg2 Si2 O6 þ MgAl2 O4 ¼ MgAl2 SiO6 þ Mg2 SiO4 Opx Spl Opx Ol (1) but olivine has not been given as a product phase. Instead of reaction (1), Liermann & Ganguly (2003) emphasized that the experimental pressure conditions of 9–14 kbar correspond closely to the maximum limiting pressures for the stability of orthopyroxene þ spinel, based on the following reaction: 4MgSiO3 þ MgAl2 O4 ¼ Mg3 Al2 Si3 O12 þ Mg2 SiO4 Opx Spl Grt Ol (2) In the present study, we report the result of reconnaissance experiments utilizing natural UHT granulite containing aluminous orthopyroxene (Al2O3 ¼ 8.31 wt%) and spinel, and propose a new Fe2þ –Mg exchange thermometer that is available for the temperature estimation of high-grade metamorphism. Mineral abbreviations are after Kretz (1983), except for Afs (alkali feldspar).
Experimental procedures We determined the Fe2þ –Mg partitioning utilizing a phlogopite-bearing orthopyroxene granulite (sample TM981229-03E; see Miyamoto et al. 2004, for
detailed sample description) from the Napier Complex in Enderby Land, East Antarctica. A brief description of the granulite is presented here. The granulite contains orthopyroxene of XMg ¼ 0.79 and spinel of XMg ¼ 0.59; the Al2O3 content of orthopyroxene is 8.3 wt% and the Cr2O3 content of spinel is 4.2 wt% (Table 1). The modal proportion of the constituent minerals is as follows: orthopyroxene 45.6 wt%; feldspar (alkali feldspar + plagioclase) 23.7%; F-bearing phlogopite 14.0%; sapphirine 12.1%; spinel 3.4%; rutile 1.2%; quartz ,1.0%. The mineral assemblage of sapphirine þ quartz is locally found at grain boundaries of granoblastic orthopyroxene in the granulite (see Miyamoto et al. 2004, fig. 5). This sapphirine þ quartz paragenesis can be regarded as evidence for UHT metamorphics at T .1030 8C (Hensen & Green 1971, 1972, 1973; Bertrand et al. 1991). All spinel grains in the granulite are overgrown by host sapphirine (see Miyamoto et al. 2004, fig. 3d; Sato et al. 2004, fig. 2). No distinctive compositional zoning in the constituent minerals has been found (Table 1). Sato et al. (2004) performed ‘phase equilibrium experiments’ utilizing the UHT granulite (TM981229-03E) at 9–13 kbar and 950–1200 8C (Table 2) to investigate the P –T phase relation for this granulite. Chemical data for run products (orthopyroxene and spinel) obtained by this series of experiments were evaluated in the present study. In the experiments, two types of starting materials were used: (1) glass, which was obtained by fusing the granulite at 10 kbar and 1670 8C for 2 min within a graphite capsule; (2) a mixture of 90 wt% glass þ 10 wt% pulverized granulite. In addition to these experiments reported by Sato et al. (2004), we performed ‘isobaric cooling experiments at 9 kbar’ utilizing the same specimen of the UHT granulite (TM981229-03E) at temperatures from 1200 8C to 900 8C at 1008C intervals (Table 2). In this series of experiments, the P–T path of isobaric cooling that characterizes the Napier Complex (Ellis 1980, 1987; Harley 1998b; Yoshimura et al. 2000) was simulated for the purpose of demonstrating the tendency of compositional variation of constituent phases under retrograde metamorphic conditions. The glass was put into a double capsule sample container with inner molybdenum-foil and outer platinum. Kawasaki & Motoyoshi (2006) mentioned that the f O2 of the charge can be kept within the Mo – MoO2 buffer and the intrinsic f O2 is 2.9 10216 bar at 1000 8C (after O’Neill 1986). The starting material was dried at 110 8C in an oven before the experiments. The edge of one side of the outer platinum tube (i.e. the outer capsule) was welded by carbon arc, and the other side was strongly crimped twice to prevent loss of vapour components
Fe2þ –Mg PARTITIONING BETWEEN Opx AND Spl
433
Table 1. Bulk composition of the UHT metamorphic granulite (TM981229-03E) and chemical composition of mafic minerals
Wt % SiO2 TiO2 Al2O3 Cr2O3 FeO† MnO MgO NiO CaO Na2O K2O BaO ZnO F O¼ H2O Total Mg/(Fe þ Mg)† Cr/(Al þ Cr)
Bulk* (n ¼ 15)
s
Opx (n ¼ 10)
s
Spl (n ¼ 10)
s
Phl (n ¼ 10)
Spr (n ¼ 10)
47.67 1.47 13.82 0.28 9.51 0.15 22.02 0.10 0.49 0.46 2.28 0.25 0.07 0.55 20.23
0.36 0.07 0.25 0.04 0.50 0.04 0.20 0.05 0.04 0.03 0.08 0.21 0.07 0.23
50.21 0.09 8.31 0.22 12.83 0.23 26.65 0.07 0.11 0.02 0.02 0.14 0.19 n.d.
0.63 0.04 0.84 0.06 0.28 0.04 0.49 0.04 0.11 0.02 0.01 0.14 0.08 n.d.
0.03 n.d. 60.53 4.20 18.11 0.14 14.70 0.62 n.d. 0.05 0.02 0.20 1.11 n.d.
0.01 n.d. 1.08 0.50 0.49 0.02 0.21 0.05 n.d. 0.03 0.02 0.20 0.10 n.d.
39.22 4.30 13.64 0.22 5.43 0.03 20.70 0.17 0.01 0.12 10.43 0.56 0.14 4.05 21.71 2.22 99.54
13.56 0.03 60.37 1.47 5.92 0.08 17.17 0.24 0.01 0.03 0.01 0.06 0.17 n.d.
98.89 0.805
99.11 0.787
99.73 0.591 0.044
0.872
99.12 0.838 0.016
*Bulk composition of the granulite was determined by defocused X-ray microprobe analysis of the glass. †Total Fe as FeO. Original data for the chemical composition in this table were presented by Sato et al. (2004). Opx, orthopyroxene; Spl, spinel; Phl, phlogopite; Spr, sapphirine (after Kretz 1983); s, standard deviation; n, number of point analyses; n.d., not determined.
during ignition. The isobaric cooling experiments were conducted based on the following procedure. Initially, four samples were kept at 9 kbar and 1200 8C for 80 h and then quenched. One of the run products (run 020311) was analysed using an electron probe micro analyser (EPMA). Other three run products were reheated at 1100 8C for 185 h (run 020320). One of them was analysed by EPMA. Two remaining samples were reheated at 1000 8C for 502 h (run 020404). The last sample was finally heated at 900 8C for 952 h (run 020614). The experiments were carried out using a 16.0 mm piston-cylinder apparatus housed at Ehime University. Graphite was used as a heater (Fig. 1). Temperature was controlled using a Pt– Pt13%Rh thermocouple (Fig. 1). Pressure was monitored by using an oil pressure gauge. The pressure was calibrated by phase transformations of Bi I –II at room temperature (25.5 kbar; Hall 1971) and by the quartz $ coesite phase transition at 1000 8C (29.7 kbar; Bohlen & Boettcher 1982). The chemical composition of all run products was analysed by EPMA (JEOL model JXA-8800 Superprobe) with the ZAF correction method. Accelerating voltage and electron beam current were 15 kV and 5.0 1029 A, respectively. Judging from the size
of contamination spots that formed as a result of chemical analysis, the beam diameter was regarded as 1– 2 mm.
Experimental results The experimental P– T conditions and results are summarized in Table 2. Crystals of orthopyroxene and spinel were found in all run products (Fig. 2 and Tables 2 and 3). The Fe2þ –Mg distribution coefficient (KD) between orthopyroxene and spinel is expressed as KD ¼
Spl Opx XFe XMg Spl Opx XMg XFe
(3)
The mean value and standard deviation of the Al2O3 content of orthopyroxene grains collected from all run products was 10.57 + 1.14 wt%. The half-reversals that utilize only glass show sometimes overshooting beyond equilibrium value (Ganguly et al. 1996). Therefore a mixture of glass and pulverized granulite was used as the starting material in our experiments, except for three
434
Table 2. Run details and Fe2þ –Mg distribution coefficients between orthopyroxene and spinel Run no.
9 9 13 12 11 9 13 12 11 9 9 13 10 9 12 11 9 9 12 9 9
T (8C)
1200 1200 1150 1150 1150 1150 1100 1100 1100 1100 1100 1050 1050 1050 1000 1000 1000 1000 950 950 900
Time (h)
80 163 161 171 172 176 119 232 262 264 185 309 340 336 575 597 598 502 770 584 952
Starting material
Gls Gls þ CM Gls þ CM Gls þ CM Gls Gls Gls þ CM Gls þ CM Gls þ CM Gls þ CM 020311 Gls þ CM Gls þ CM Gls þ CM Gls þ CM Gls þ CM Gls þ CM 020320 Gls þ CM Gls þ CM 020404
Final XMg Opx
Spl
0.868 0.857 0.806 0.818 0.860 0.835 0.801 0.808 0.808 0.828 0.916 0.802 0.817 0.816 0.809 0.800 0.803 0.849 0.820 0.809 0.841
0.813 0.805 0.714 0.737 0.784 0.752 0.688 0.695 0.691 0.722 0.859 0.684 0.698 0.691 0.671 0.677 0.673 0.739 0.653 0.653 0.703
Final KD (s)
Additional phases and XAv. Mg
1.51 (0.02) 1.45 (0.04) 1.66 (0.04) 1.60 (0.05) 1.69 (0.04) 1.67 (0.03) 1.83 (0.04) 1.85 (0.05) 1.88 (0.03) 1.85 (0.06) 1.78† 1.87 (0.03) 1.93 (0.05) 1.98 (0.06) 2.08 (0.04) 1.91 (0.06) 1.98 (0.06) 1.99 (0.07) 2.42 (0.06) 2.25 (0.08) 2.24 (0.07)
Melt (0.72) Melt (0.69) Phl (0.85), Spr (0.84), melt (0.54), + Rt Phl (0.86), melt (0.59) Phl (0.89), melt (0.61) Phl (0.89), melt (0.52) Afs, Phl (0.85), Spr (0.84), Rt Afs, Phl (0.85), Spr (0.83), Rt Afs, Phl (0.85), Spr (0.83), Rt Phl (0.87), Spr (0.84), Rt, melt (0.58) Phl (0.93), Rt, melt (0.63) Afs, Pl, Phl (0.86), Spr (0.83), Rt, Grt (0.69) Afs, Pl, Phl (0.85), Spr (0.84), Rt Afs, Pl, Phl (0.85), Spr (0.84), Rt Afs, Pl, Phl (0.85), Spr (0.83), Rt, Grt (0.68) Afs, Pl, Phl (0.86), Spr (0.84), Rt Afs, Pl, Phl (0.85), Spr (0.83), Rt Phl (0.87), Rt, melt (0.61) Afs, Pl, Phl (0.84), Spr (0.83), Rt, Grt (0.69) Afs, Pl, Phl (0.85), Spr (0.84), Rt, + Qtz Phl (0.88), Rt, melt (0.54)
*The isobaric cooling experiment was performed at 9 kbar and temperatures from 1200 to 900 8C at 100 8C intervals. †The KD value of run 020320 was almost identical to those of other runs at 1100 8C. However, the XMg values of orthopyroxene and spinel of run 020320 deviated remarkably from others. Thus, this KD data were omitted from a calibration of the exchange thermometer. Afs, Alkali feldspar; Pl, plagioclase; Rt, rutile; Qtz, quartz (after Kretz 1983, except for Afs); Gls, glass; Gls þ CM, glass 90 wt% þ CM 10 wt% (Gls, glass; CM, crystalline mixture of Opx, Spr, Spl, Afs, Pl, Phl, Rt, +Qtz).
K. SATO ET AL.
020311 030924A 021022B 030613B 020920D 020902C 031001C 031020B 030526B 030804A 020320* 021029C 030402C 021111A 031030B 030421C 030623B 020404* 031123B 021127A 020614*
P (kbar)
Fe2þ –Mg PARTITIONING BETWEEN Opx AND Spl
435
Fig. 1. Sectional view of a typical cell assembly that was used in the experiments.
runs (i.e. 020311, 020902C and 020920D; Table 2). As a result of the experiments, compositional zoning in synthetic orthopyroxene was found, with Fe-rich core and Mg-rich rim, although the runs were 119 –770 h long. The XMg of synthetic spinel also varied, but because of the small grain size (, 10 mm) of spinel, we were unable to determine the trend of compositional zoning in these Opx Spl and XMg in the run grains (e.g. variation of XMg product (030804A); see Fig. 3). If a glass or a mixture of glass þ seed minerals is used as a starting material, the compositional zoning in synthetic phases is commonly found in run products (e.g. Harley 1984; Kawasaki & Sato 2002; Sato & Kawasaki 2002). For two different experimental techniques (i.e. phase equilibrium experiments
Fig. 2. Back-scattered electron image of run product that was obtained at 9 kbar and 1100 8C (run 030804A). The mineral phases orthopyroxene (Opx) þ spinel (Spl) þ F-phlogopite (Phl) þ sapphirine (Spr) þ rutile (Rt) were found coexisting with melt. (Note the presence of euhedral spinel crystal.)
and isobaric cooling experiments), the synthetic orthopyroxene and spinels that can be regarded to be in equilibrium or near equilibrium were selected as described below. In the case of ‘phase equilibrium experiments’, the compositional zoning of orthopyroxene might depend on the kinetics of nucleation and growth of minerals. The experiments were performed in the fixed composition that can be approximated to aluminous orthopyroxene (Table 1). The Mg#Bulk Opx and XMg in the granulite (TM981229-03E) are 0.81 and 0.79, respectively (Table 1). In this bulk composition, orthopyroxene can grow at the expense of the Mg-rich glass. Therefore, the rim of orthopyroxene becomes Mg-rich. Consequently, the Mg#Melt should shift to the more Fe-rich side. In most cases in our experiments, the seed mineral Spl ¼ 0.59) was aggregate containing spinel (XMg added to the glass (Table 2). In such runs, we regarded Mg-rich spinel as final grains that were produced by the experiments because the XMg values of Spl in Seed synthetic spinels are always higher than XMg (Fig. 3 and Table 2). On the other hand, for the three runs (020311, 020920D and 020902C) using seed-free starting material, spinel was formed only from the glass. In these three products, a compositional relationship between spinel and melt was Spl . Mg#Melt (Table 2). Therealways found as XMg fore, the mineral pairs of Mg-rich orthopyroxene and Mg-rich spinel in the ‘phase equilibrium experiments’ were adopted for the determination of the Fe2þ –Mg exchange thermometer (Table 2). The Fe2þ –Mg partitioning between orthopyroxene and spinel was also investigated by ‘isobaric cooling experiments’ at 9 kbar. At 1200 8C, the phase assemblage of orthopyroxene þ spinel þ melt was obtained as a primary assembly for this series of experiments (run 020311; Table 2). At lower temperatures of 1100–900 8C, the phase assemblage of orthopyroxene þ spinel þ F-bearing phlogopite þ rutile þ melt was
436
K. SATO ET AL.
Table 3. Chemical composition of orthopyroxene and spinel in run products (Table 3 continued on pp’s 437, 438, 439 & 440.) Run no.: P, T:
020311 9 kbar, 1200 8C
030924A 9 kbar, 1200 8C
021022B 13 kbar, 1150 8C
030613B 12 kbar, 1150 8C
Phase:
Opx
Spl
Opx
Spl
Opx
Spl
Opx
Spl
wt% SiO2 TiO2 Al2O3 Cr2O3 FeO* MnO MgO NiO CaO Na2O K2O Total
51.17 1.01 8.79 0.23 8.30 0.21 29.97 0.01 0.19 0.01 0.01 99.90
0.90 0.62 64.45 3.02 8.82 0.14 21.55 0.06 0.00 0.05 0.09 99.70
49.98 1.20 9.63 0.16 7.95 0.11 26.78 0.11 0.35 0.08 0.78 97.13
0.94 0.54 62.73 2.78 8.96 0.18 20.76 0.08 0.00 0.04 0.09 97.10
48.95 0.95 11.10 0.30 11.18 0.28 26.01 0.18 0.34 0.03 0.06 99.38
0.52 0.43 63.53 2.52 13.70 0.18 18.79 0.25 0.09 0.07 0.06 100.14
47.85 1.15 12.24 0.33 10.19 0.25 25.69 0.25 0.27 0.06 0.06 98.34
0.22 0.48 63.74 2.24 11.98 0.11 18.84 0.22 0.04 0.03 0.05 97.95
Oxygen Si Ti Al Cr Fe3þ Fe2þ Mn Mg Ni Ca Na K
6 1.787 0.027 0.362 0.006 0.005 0.237 0.006 1.561 0.000 0.007 0.001 0.000
4 0.023 0.012 1.905 0.060 0.000 0.185 0.003 0.806 0.001 0.000 0.002 0.003
6 1.807 0.033 0.410 0.005 0.000 0.240 0.003 1.443 0.003 0.014 0.006 0.036
4 0.024 0.010 1.907 0.057 0.000 0.193 0.004 0.795 0.002 0.000 0.002 0.003
6 1.748 0.026 0.467 0.008 0.000 0.334 0.008 1.385 0.005 0.013 0.002 0.003
4 0.013 0.008 1.906 0.051 0.006 0.286 0.004 0.713 0.005 0.002 0.003 0.002
6 1.723 0.031 0.519 0.009 0.000 0.307 0.008 1.379 0.007 0.010 0.004 0.003
4 0.006 0.009 1.943 0.046 0.000 0.259 0.002 0.726 0.005 0.001 0.002 0.002
Mg/(Fe þ Mg)* Mg/(Fe2þ þ Mg) Fe3þ/(Fe3þ þ Fe2þ) Cr/(Al þ Cr þ Fe3þ)
0.866 0.868 0.021
0.813 0.813 0.000 0.030
0.857 0.857 0.000
0.805 0.805 0.000 0.029
0.806 0.806 0.000
0.710 0.714 0.019 0.026
0.818 0.818 0.000
0.737 0.737 0.000 0.023
KD (Fe2þ –Mg)
1.51
1.45
1.66
1.60
*Total Fe as FeO.
Fig. 3. Histogram of XMg values of orthopyroxene and spinel that were obtained at 9 kbar and 1100 8C (run 030804A). A mixture of glass þ seed minerals was used as the starting material in this run.
found (runs 020320, 020404 and 020614; Table 2). For this series of experiments, because the KD between orthopyroxene and spinel should increase with decreasing temperature, the mineral pairs of Mg-rich orthopyroxene and Fe-rich spinel were adopted for the geothermometry. As Ganguly & Tirone (1999) pointed out, the correct selection of chemical data for synthetic minerals is an important factor when geothermometry utilizing the Fe2þ – Mg distribution coefficient (KD) is performed; thus chemical data for very small mineral crystals are unsuited for the determination of the geothermometer and were omitted from dataset. The relation between ln KD and 1/T can be given (Fig. 4), based on the experimental data for KD summarized in Table 2.
Fe2þ –Mg PARTITIONING BETWEEN Opx AND Spl
437
Table 3. Continued 020920D 11 kbar, 1150 8C
020902C 9 kbar, 1150 8C
031001C 13 kbar, 1100 8C
031020B 12 kbar, 1100 8C
Opx
Spl
Opx
Spl
Opx
Spl
Opx
Spl
49.44 0.95 10.77 0.26 9.10 0.23 28.68 0.04 0.12 0.01 0.02 99.62
0.54 0.49 63.53 2.86 10.26 0.10 20.83 0.08 0.04 0.01 0.08 98.82
48.72 0.90 10.49 0.30 10.66 0.19 27.38 0.11 0.15 0.00 0.04 98.94
0.35 0.36 62.42 2.94 12.06 0.13 19.43 0.07 0.04 0.05 0.16 98.01
47.97 0.76 10.97 0.31 10.95 0.29 24.74 0.13 0.34 0.09 0.01 96.56
0.00 0.30 63.97 2.60 13.68 0.14 16.89 0.45 0.00 0.02 0.00 98.05
48.64 0.85 11.76 0.32 10.56 0.17 24.93 0.02 0.27 0.05 0.25 97.82
0.07 0.35 63.35 2.40 13.88 0.00 17.76 0.07 0.00 0.02 0.03 97.93
6 1.737 0.025 0.446 0.007 0.023 0.244 0.007 1.503 0.001 0.005 0.001 0.001
4 0.014 0.009 1.904 0.057 0.000 0.218 0.002 0.790 0.002 0.001 0.000 0.003
6 1.737 0.024 0.441 0.008 0.030 0.288 0.006 1.455 0.003 0.006 0.000 0.002
4 0.009 0.007 1.901 0.060 0.014 0.246 0.003 0.749 0.001 0.001 0.003 0.005
6 1.766 0.021 0.476 0.009 0.000 0.337 0.009 1.358 0.004 0.013 0.006 0.000
4 0.000 0.006 1.970 0.054 0.000 0.299 0.003 0.658 0.009 0.000 0.001 0.000
6 1.765 0.023 0.503 0.009 0.000 0.320 0.005 1.348 0.001 0.010 0.004 0.012
4 0.002 0.007 1.946 0.049 0.000 0.303 0.000 0.690 0.001 0.000 0.000 0.000
0.849 0.860 0.086
0.784 0.784 0.000 0.029
0.821 0.835 0.094
0.742 0.752 0.054 0.030
0.801 0.801 0.000
0.688 0.688 0.000 0.027
0.808 0.808 0.000
0.695 0.695 0.000 0.025
1.69
1.67
1.83
Thermodynamic background 2þ
For the Fe – Mg exchange reaction between orthopyroxene and spinel, 1=2Fe2 Si2 O6 þ MgAl2 O4 ¼ 1=2Mg2 Si2 O6 Spl Opx Opx þ FeAl2 O4 Spl
(4)
1.85
assuming that the volume change DV8 is independent of P kbar. DH8 and DS8 are the enthalpy change and entropy change of the exchange reaction (4), respectively. R is the gas constant (¼8.3144 J mol21 K21 ¼ 1.9862 cal mol21K21) and K is the equilibrium constant. If DV8 is independent of temperature, equation (5) can be rewritten as DH1 bar,T þ TDS1 bar,T W
W
(P kbar 1bar)DV1bar,298 K W
the free energy change DG8 at the standard state referring to the pure phases at pressure P (kbar) and temperature T (K) of interest is given as DGP,T ¼ DG1bar,T (Pkbar 1bar)DV1bar,T W
W
W
¼ DH1bar,T þ TDS1bar,T W
W
(P kbar 1 bar)DV1bar,T ¼ RT ln K
(5)
¼ RT ln K
(6)
The equilibrium constant K of equation (6) is defined by the quotient of activities: 1=2 Opx aSpl a FeAl2 O4 Mg2 Si2 O6 (7) K¼ 1=2 : Spl aMgAl2 O4 aOpx Fe2 Si2 O6
438
K. SATO ET AL.
Table 3. Continued Run no.: P, T:
030526B 11 kbar, 1100 8C
030804A 9 kbar, 1100 8C
021029C 13 kbar, 1050 8C
030402C 10 kbar, 1050 8C
Phase:
Opx
Spl
Opx
Spl
Opx
Spl
Opx
Spl
wt% SiO2 TiO2 Al2O3 Cr2O3 FeO* MnO MgO NiO CaO Na2O K2O Total
48.78 1.15 10.80 0.20 11.02 0.13 25.64 0.06 0.28 0.11 0.56 98.73
0.09 0.50 63.73 3.18 14.18 0.13 17.75 0.23 0.00 0.03 0.01 99.83
50.29 0.77 10.05 0.12 11.18 0.19 28.04 0.10 0.15 0.00 0.00 100.89
0.14 0.38 63.95 3.24 12.37 0.19 18.00 0.33 0.01 0.02 0.04 98.67
49.14 0.72 11.68 0.38 11.55 0.15 26.16 0.21 0.28 0.09 0.04 100.40
0.90 0.29 61.94 3.31 14.02 0.12 17.03 0.41 0.00 0.03 0.16 98.21
50.51 0.79 10.09 0.25 10.64 0.16 26.68 0.28 0.28 0.09 0.02 99.79
0.21 0.29 63.71 3.55 13.93 0.11 18.05 0.00 0.00 0.00 0.00 99.85
Oxygen Si Ti Al Cr Fe3þ Fe2þ Mn Mg Ni Ca Na K
6 1.752 0.031 0.457 0.006 0.004 0.327 0.004 1.373 0.002 0.011 0.008 0.026
4 0.002 0.010 1.930 0.065 0.000 0.305 0.003 0.680 0.005 0.000 0.001 0.000
6 1.759 0.020 0.414 0.003 0.024 0.303 0.006 1.462 0.003 0.006 0.000 0.000
4 0.004 0.007 1.948 0.066 0.000 0.267 0.004 0.694 0.007 0.000 0.001 0.001
6 1.736 0.019 0.486 0.011 0.001 0.340 0.004 1.378 0.006 0.011 0.006 0.002
4 0.024 0.006 1.912 0.069 0.000 0.307 0.003 0.665 0.009 0.000 0.002 0.005
6 1.792 0.021 0.422 0.007 0.000 0.316 0.005 1.411 0.008 0.011 0.006 0.001
4 0.005 0.006 1.926 0.072 0.000 0.299 0.002 0.690 0.000 0.000 0.000 0.000
Mg/(Fe þ Mg)* Mg/(Fe2þ þ Mg) Fe3þ/(Fe3þ þ Fe2þ) Cr/(Al þ Cr þ Fe3þ)
0.806 0.808 0.012
0.691 0.691 0.000 0.032
0.817 0.828 0.073
0.722 0.722 0.000 0.033
0.801 0.802 0.003
0.684 0.684 0.000 0.035
0.817 0.817 0.000
0.698 0.698 0.000 0.036
KD (Fe2þ 2 Mg)
1.88
1.85
If the orthopyroxene and spinel are not an ideal solid solution (a ¼ Xg, where g is the activity coefficient), then K should be distinguished from the KD: K¼
Spl Opx Spl Opx XFe XMg gFeAl2 O4 gMgSiO3 ¼ KD Kg : Spl Opx Opx XMg XFe gSpl MgAl2 O4 gFeSiO3
(8)
Thus, for the Fe2þ –Mg exchange reaction (4), the following equation can be derived from equation (6): DH1 bar,T þ TDS1 bar,T (P kbar 1 bar)DV1 bar,298K W
W
¼ RT ln KD þ RT ln Kg :
W
(9)
The thermodynamic parameters DH8, DS8 and DV8 can be regarded as constant or nearly constant for the range of P– T conditions of our experiments. RT ln Kg in equation (9) is a function of T and
1.87
1.93
composition. In the experiments, the bulk composition of the starting materials is constant, and Spl variations of XOpx Mg and XMg in the run products range between 0.800 and 0.868, and 0.653 and 0.813, respectively (Table 2). Therefore, in the present geothermometry, RT ln Kg in equation (9) was assumed to be constant or nearly constant (Kawasaki & Matsui 1983; Lee & Ganguly 1988; Liermann & Ganguly 2003), and was neglected. Hence, RT ln KD can be approximated as DH1bar,T þ TDS1bar,T W
(P kbar 1bar)DV1bar,298 K W
¼ RT lnKD
(10)
where DS* (¼ DS8 2 R ln Kg) is the effective entropy change.
Fe2þ –Mg PARTITIONING BETWEEN Opx AND Spl
439
Table 3. Continued 021111A 9 kbar, 1050 8C
031030B 12 kbar, 1000 8C
030421C 11 kbar, 1000 8C
030623B 9 kbar, 1000 8C
Opx
Spl
Opx
Spl
Opx
Spl
Opx
Spl
48.74 0.71 11.32 0.27 11.18 0.23 26.50 0.09 0.24 0.07 0.03 99.38
0.04 0.16 65.70 3.30 14.23 0.09 17.82 0.29 0.00 0.06 0.01 101.70
51.16 0.62 9.61 0.31 11.98 0.22 27.42 0.18 0.33 0.00 0.16 101.99
0.00 0.20 64.59 2.99 14.83 0.06 16.94 0.62 0.00 0.04 0.04 100.31
49.85 0.81 9.84 0.24 11.42 0.20 25.62 0.04 0.31 0.12 0.58 99.03
0.08 0.11 62.31 3.29 14.13 0.07 16.58 0.38 0.00 0.03 0.04 97.02
49.56 0.58 10.90 0.27 11.43 0.16 26.12 0.00 0.28 0.05 0.01 99.36
0.13 0.10 64.62 2.94 14.93 0.06 17.24 0.43 0.00 0.04 0.01 100.50
6 1.735 0.019 0.475 0.008 0.016 0.317 0.007 1.406 0.003 0.009 0.005 0.001
4 0.001 0.003 1.950 0.066 0.000 0.300 0.002 0.669 0.006 0.000 0.003 0.000
6 1.780 0.016 0.394 0.009 0.013 0.336 0.006 1.422 0.005 0.012 0.000 0.007
4 0.000 0.004 1.952 0.061 0.000 0.318 0.001 0.648 0.013 0.000 0.002 0.001
6 1.788 0.022 0.416 0.007 0.000 0.343 0.006 1.370 0.001 0.012 0.008 0.027
4 0.002 0.002 1.946 0.069 0.000 0.313 0.002 0.655 0.008 0.000 0.002 0.001
6 1.768 0.016 0.458 0.008 0.000 0.341 0.005 1.389 0.000 0.011 0.003 0.000
4 0.003 0.002 1.947 0.059 0.000 0.319 0.001 0.657 0.009 0.000 0.002 0.000
0.809 0.816 0.048
0.691 0.691 0.000 0.033
0.803 0.809 0.036
0.671 0.671 0.000 0.030
0.800 0.800 0.000
0.677 0.677 0.000 0.034
0.803 0.803 0.000
0.673 0.673 0.000 0.030
1.98
2.08
1.91
Opx In our experiments, XAl (¼ mole fraction of Al2O3/(Al2O3 þ MgSiO3 þ FeSiO3)Opx) is limited Opx to about 0.036 –0.044. Because variation of XAl is small, we can assume that the influence of Al content of orthopyroxene on KD is constant. Even Opx is applied to the previous empirical if this XAl equation of Liermann & Ganguly (2003) utilizing thermodynamic data (see equation (8) and the third column of table 4 of Liermann & Ganguly 2003), the influence is only about 1%. As a rough tendency, XCr increases with decreasing XMg (Fig. 5), which agrees with the notion that spinel is a reciprocal solid solution in the Cr-bearing system (e.g. Wood & Nicholls 1978). Therefore, the effect of reciprocal exchange reaction of spinel,
FeCr2 O4 þ MgAl2 O4 ¼ MgCr2 O4 þ FeAl2 O4
(11)
1.98
is important for the determination of the orthopyroxene –spinel exchange thermometer (Liermann & Ganguly 2003). However, the XCr values of the synthetic spinels have only the range 0.023–0.038 (Fig. 5). Liermann & Ganguly (2003) investigated the effect of XCr variation in spinel at 1000 8C. If the XCr variation of synthetic spinels obtained in our experiments is applied to Liermann & Ganguly’s empirical equation, the effect is estimated to be about 3% when KD is 2.00 (around 1000 8C). Therefore, the effects of variation in Opx Spl and XCr on KD were neglected in the XAl present geothermometry. For the Fe2þ –Mg exchange reaction between orthopyroxene and spinel of equation (4), DV8 can be calculated as 214.90 cal kbar21 (Table 4) on the basis of the thermodynamic dataset for orthopyroxene (Matsui et al. 1968) and spinel (Robie et al. 1978). In our experiments, the pressures
440
K. SATO ET AL.
Table 3. Continued Run no.: P, T:
020404 9 kbar, 1000 8C
031123B 12 kbar, 950 8C
021127A 9 kbar, 950 8C
020614 9 kbar, 900 8C
Phase:
Opx
Spl
Opx
Spl
Opx
Spl
Opx
Spl
wt% SiO2 TiO2 Al2O3 Cr2O3 FeO* MnO MgO NiO CaO Na2O K2O Total
48.76 1.01 9.15 0.19 9.48 0.26 27.85 0.05 0.15 0.05 0.00 96.95
0.26 0.16 64.96 2.59 11.55 0.22 18.39 0.00 0.03 0.04 0.02 98.22
49.29 0.60 11.49 0.22 11.97 0.18 26.90 0.11 0.19 0.06 0.07 101.08
0.29 0.20 63.12 3.39 15.54 0.17 16.42 0.54 0.05 0.02 0.01 99.75
48.82 0.56 8.51 0.20 11.95 0.15 26.46 0.10 0.12 0.01 0.05 96.93
0.13 0.22 62.41 3.41 15.59 0.12 16.49 0.51 0.02 0.04 0.04 98.98
50.65 1.09 9.65 0.26 9.55 0.17 28.36 0.11 0.13 0.05 0.01 100.03
0.18 0.13 66.44 2.60 13.49 0.16 17.89 0.02 0.00 0.00 0.05 100.96
Oxygen Si Ti Al Cr Fe3þ Fe2þ Mn Mg Ni Ca Na K
6 1.766 0.028 0.391 0.005 0.020 0.268 0.008 1.504 0.001 0.006 0.004 0.000
4 0.007 0.003 1.973 0.053 0.000 0.249 0.005 0.707 0.000 0.001 0.002 0.001
6 1.726 0.016 0.474 0.006 0.043 0.308 0.005 1.404 0.003 0.007 0.004 0.003
4 0.008 0.004 1.929 0.070 0.000 0.337 0.004 0.635 0.011 0.001 0.001 0.000
6 1.787 0.015 0.367 0.006 0.026 0.340 0.005 1.444 0.003 0.005 0.001 0.002
4 0.003 0.004 1.922 0.070 0.000 0.341 0.003 0.642 0.011 0.001 0.002 0.001
6 1.780 0.029 0.400 0.007 0.000 0.281 0.005 1.486 0.003 0.005 0.003 0.000
4 0.005 0.002 1.977 0.052 0.000 0.285 0.003 0.673 0.000 0.000 0.000 0.002
Mg/(Fe þ Mg)* Mg/(Fe2þ þ Mg) Fe3þ/(Fe3þ þ Fe2þ) Cr/(Al þ Cr þ Fe3þ)
0.840 0.849 0.068
0.739 0.739 0.000 0.026
0.800 0.820 0.123
0.653 0.653 0.000 0.035
0.798 0.809 0.070
0.653 0.653 0.000 0.035
0.841 0.841 0.000
0.703 0.703 0.000 0.026
KD (Fe2þ –Mg)
1.99
2.42
2.25
2.24
*Total Fe as FeO.
range from 9 to 13 kbar. Considering the negligible pressure effect, an average of 11 kbar is adopted. Thus, equation (10) can be rewritten as DH1 bar,T þ TDS1 bar,T W
þ 14:90(11 kbar 1 bar) ¼ RT ln KD
(12)
If we assumed a negligible pressure effect in equation (12), this equation can be modified as þ TDS11kbar,T ¼ RT ln KD11 kbar,T DH11kbar,T
(Deming 1943) stands for DH* ¼ 24650 + 334 cal and DS* ¼ 22.2524 + 0.079 cal K21 (Fig. 4). From these data, the Fe2þ –Mg distribution coefficient (KD) between orthopyroxene and spinel can be approximated as follows: ln KD ¼ (1:134 + 0:040) þ [(2341 + 168) þ 7:5(P 11)]=T:
(14)
The effect of the pressure is also small (Fig. 6). If this effect is neglected (P ¼ 11 kbar), we can rewrite equation (14) as
(13)
ln KD ¼ (1:134 + 0:040) þ (2341 + 168)=T: (15)
where DH* is DH8 þ (P kbar 21 bar)DV8. The linear fit utilizing the least-squares method
This approach for geothermometry has been supported by previous studies (Sakai & Kawasaki
Fe2þ –Mg PARTITIONING BETWEEN Opx AND Spl
441
Fig. 4. Relation between Fe2þ – Mg distribution coefficient KD and temperature T (K). The continuous line was approximated from all the symbols (W, †, 4). Chemical data of the run 031123B were excluded from the figure because orthopyroxene in this run product has a high Fe3þ/(Fe3þ þ Fe2þ) value. For comparison, the previous experimental result (Liermann & Ganguly 2003, fig. 3a) is shown, as a dashed line.
1997; Kawasaki & Motoyoshi 2000; Kawasaki & Sato 2002; Sato & Kawasaki 2002; Sato et al. 2006).
Discussion and conclusions
Fig. 5. Relation between XMg and XCr of spinel in run products and natural UHT metamorphic rocks that were selected from the Napier Complex. The chemical data for spinel from natural aluminous gneisses are for samples 9812280101 and 9812290503 of Yoshimura et al. (2000).
In this study, equation (14) (or equation (15)) is given as a new exchange thermometer using the Fe2þ –Mg partitioning between aluminous orthopyroxene and spinel. Liermann & Ganguly (2003) found a temperature dependence of the Fe2þ –Mg reaction between aluminous orthopyroxene and spinel. We report a higher temperature dependence when compared with the finding of Liermann & Ganguly (2003). This exchange reaction has the potential to be a highly sensitive thermometer. The new thermometer was applied to high-grade metamorphic rocks including the UHT rocks from the Napier Complex, Lu¨tzow-Holm Complex and Rayner Complex in East Antarctica, and other selected metamorphic terranes of the world (Eastern Ghats and Madurai block in India, Central Highland Complex in Sri Lanka, South Harris in NW Scotland, In Ouzzal Terrane in Algeria, Epupa Complex in Namibia, Andriamena
442
K. SATO ET AL.
Table 4. Volume data for orthopyroxene and spinel End member MgSiO3 FeSiO3 MgAl2O4 FeAl2O4
V8 (cal kbar21 mol21)*
Reference
749.08 788.84 949.09 973.95
Matsui et al. (1968) Matsui et al. (1968) Robie et al. (1978) Robie et al. (1978)
*For the Fe2þ –Mg exchange reaction FeSiO3 þ MgAl2O4 ¼ MgSiO3 þ FeAl2O4, DV8 is –14.90 cal kbar21.
Mafic Unit in Madagascar and western –central Opx Spl and XCr Kyushu in Japan; Table 5). The XAl values for the selected metamorphic rocks range from 0.004 to 0.038 and up to 0.153, respectively. Opx In the present geothermometry, the effects of XAl Spl and XCr were neglected. For these metamorphic rocks, our proposed temperature varies between 653 and 902 8C. Temperature estimation by the new thermometer could not give the temperature conditions for UHT metamorphism, except for the Andriamena Mafic Unit (Table 5). The effect of pressure change on the Fe2þ –Mg exchange reaction between orthopyroxene and spinel was very small (3 8C or 4 8C per kbar; see equation (14) and Fig. 6). The illustrated comparison of estimated temperatures between the present and two previous Fe2þ –Mg exchange thermometers in Figure 7 shows that each thermometer is in good agreement with the others on temperature estimation. We also tried to apply the two exchange thermometers to a UHP metamorphic rock
Fig. 6. Pressure effect for the present orthopyroxene– spinel exchange geothermometer. Continuous lines: pressure effect was estimated using the thermodynamic data of Matsui et al. (1968) and Robie et al. (1978); dashed lines: the effect was ignored.
containing orthopyroxene (Al2O3-free) and Cr-spinel (XCr ¼ 0.555) from Dabie Shan in China (Okay 1994; Table 5). The KD between this orthopyroxene and spinel is remarkably high (26.70), and the pressure (28 –40 kbar) is also high, as compared with those of the present experiments. The temperatures estimated for this UHP metamorphic rock by our geothermometer and that of Liermann & Ganguly (2003) were 286–306 8C and 513– 554 8C, respectively. This effect was the result of the property that the estimated temperature increases with increasing pressure and decreases with increasing KD (see equation (14)), and the effects Opx Spl and XCr , which are described below. of XAl Our geothermometer has no correction for the Opx Spl and XCr . According to the correceffects of XAl tion proposed by Liermann & Ganguly (2003), if
Fig. 7. Comparison of estimated temperatures between three Fe2þ – Mg exchange thermometers. S06: sapphirine–spinel thermometer of Sato et al. (2006). KS02: sapphirine– orthopyroxene thermometer of Kawasaki & Sato (2002). The pressure condition was fixed at 11 kbar for these estimations. Data sources of mineral pairs for the estimations are given in Table 5.
Table 5. Comparison of estimated temperatures (8C) of various UHT and/or high-grade metamorphic rocks Area
Opx XAl
Spl XCr
7–13* 7–13*
0.031 0.032
10–12*
KD
Opx – Spl Present T
Opx –Spl TLG03
Opx– Grt TKM00
Opx – Grt TG96
Opx– Spr TKS02
Spr – Spl TD06
0.044 0.024
2.56 2.43
841 – 863 870 – 892
772 – 829 776 – 835
– 832
– 745– 780
912 946
947 927
873 – 895 913 – 936
0.026
0.005
2.57
850 – 857
733 – 752
928
878– 891
839
986
817 – 824
7–12
0.032
–
2.58
837 – 855
684 – 731
1014
965– 1001
1083
925
917 – 936
9 8
0.038 0.004
– –
2.72 3.21
817 735
657 589
– 699
– 601
947 –
917 –
–
9–11
0.027
0.001
2.43
877 – 885
759 – 779
906
845– 858
995
878
1041 – 1050
9
0.027
0.001
2.69
822
682
844
771
1174
925
926
9–14
0.028
0.036
2.96
775 – 792
671 – 713
772
687– 714
701
908
964 – 984
5–10
0.016
0.153
3.84
653 – 668
682 – 718
608
490– 512
640
1022
745 – 762
5–6
0.019
0.007
2.51
845 – 848
718 – 727
839
742– 748
–
–
0.029
0.002
2.30
902
787
803
711
747
971
830
7–8
0.012
–
2.35
889 – 893
791 – 801
–
–
–
995
792 – 795
28–40
–
0.555
26.70
286 – 306
513 – 554
693
694– 753
–
–
7
Spr – Spl TS06
943
–
Fe2þ –Mg PARTITIONING BETWEEN Opx AND Spl
East Antarctica Napier Complex Howard Hills1 Tonagh Island2 Lu¨tzow-Holm Complex Rundva˚gshetta3 Rayner Complex Forefinger Point4 India Eastern Ghats Sunkarametta5 Kondapalle6 Madurai block Ganguvarpatti7 Sri Lanka Central Highland Complex8 South Harris, Scotland Lewisian Complex9 NW Hoggar, Algeria In Ouzzal Terrane10 NW Namibia Epupa Complex11 North–central Madagascar Andriamena Mafic Unit12 Kyushu, Japan Higo metamorphic terrane13 China Dabie Shan (UHP rock)14
P (kbar)
–
443
*Sources of pressure range: Howard Hills and Tonagh Island, Harley (1998a); Rundva˚gshetta, Motoyoshi & Ishikawa (1997). Present estimations (the new Opx – Spl thermometer): ln KD ¼ 2 1.134 þ [2341 þ 7.5(P 2 11)]/T. Other estimations: TLG03, Opx – Spl thermometer (Liermann & Ganguly 2003, including effects of charge balance and AlOpx); TKM00, Opx –Grt thermometer (Kawasaki & Motoyoshi 2000); TG96, Opx –Grt thermometer (Ganguly et al. 1996); TKS02, Opx –Spr thermometer (Kawasaki & Sato 2002); TD06, Spr–Spl thermometer (Das et al. 2006); TS06, Spr –Spl thermometer (Sato et al. 2006). Sources of data, and sample number: 1Sato et al. (2004), TM981229-03E; 2Hokada et al. (1999), A98021102H; 3Kawasaki et al. (1993), 92011102A; 4Harley et al. (1990), 4652; 5Bose et al. (2000), bS1/1 (S5/1A); 6Sengupta et al. (1999), K264C; 7Sajeev et al. (2004), Spl –Opx; 8Sajeev & Osanai (2004), Type C (Opx –Spl and Opx –Spr); 9Baba (2003), 95919-18; 10Ouzegane et al. (2003), lnh Opx is the mole fraction of Al2O3 in Opx [¼ Al2O3/ 131; 11Brandt et al. (2003), 458-3; 12Goncalves et al. (2004), C43; 13Osanai et al. (1998), Assemblages 1 and 5; 14Okay (1994), 568G. XAl Spl (Al2O3 þ MgSiO3 þ FeSiO3)Opx]. XCr is Cr/(Al þ Cr)Spl or Cr/(Al þ Cr þ Fe3þ)Spl.
444
K. SATO ET AL.
Opx Spl Table 6. Estimated temperatures (8C) and effects of XAl and XCr based on geothermometer of Liermann & Ganguly (2003) Opx XAl
0.00 0.01 0.02 0.03 0.04
Spl XCr
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.10
774 759 745 730 716
792 777 763 748 734
810 796 781 767 752
828 814 799 785 770
846 832 817 803 789
864 850 836 821 807
883 868 854 839 825
901 886 872 857 843
919 905 890 876 861
937 923 908 894 879
955 941 926 912 898
See Eq. (8) and the third column of table 4 of Liermann & Ganguly (2003). KD and P were fixed for the estimations, at 2.56 and 11 kbar, respectively.
KD and the pressure are given as 2.56 and 11 kbar, the estimated temperature changes from 846 – Opx ¼ 0.00 –0.04 and invari789 8C at variable XAl Spl Opx effect able XCr ¼ 0.04 (Table 6). Thus, the XAl in Liermann & Ganguly’s thermometer is given as Opx 0:03) þ 803 T( C) ¼ 1425(XAl W
Spl : constant): (KD , P, XCr
(16)
On the other hand, the temperature changes Opx from 730 –912 8C at invariable XAl ¼ 0.03 and Spl variable XCr ¼ 0.00–0.10 (KD ¼ 2.56 and Spl effect in P ¼ 11 kbar; Table 6). Thus, the XCr Liermann & Ganguly’s thermometer is given as Spl T( C) ¼ 1820(XCr 0:04) þ 803 W
Opx : constant): (KD , P, XAl
(17)
In the case of temperature estimation for the UHP metamorphic rock from Dabie Shan, there is a large difference between the estimated temperatures of 286– 306 8C and 513 –554 8C (Table 5). Spl value This is mainly caused by the high XCr (0.555). Therefore, the new thermometer cannot Spl , especially the be applied for rocks with high XCr Dabie Shan UHP metamorphic rock. Our geothermometer does not yield UHT metamorphic conditions (Table 5), as mentioned above, because it is relatively easy for spinel to re-equilibrate during the cooling stage of metamorphism. Hence, temperatures derived by the spinel-bearing Fe2þ –Mg exchange thermometers may not correspond to the thermal peak of UHT metamorphism. Thus, it is suggested that such temperatures are given as the closure temperatures of Fe2þ –Mg partitioning during the cooling stage of UHT metamorphism and/or the closure
temperatures on the near-peak conditions of granulite-facies metamorphism. In fact, the estimated temperatures are in good agreement with the retrograde metamorphic conditions in the Napier Complex (e.g. Yoshimura et al. 2000; Kawasaki et al. 2002; Sato et al. 2006; Table 5) and with the near thermal peak in the Lu¨tzow-Holm Complex (e.g. Kawasaki & Motoyoshi 2006). The authors received invaluable discussion and suggestions from S. L. Harley, K. Shiraishi, E. Takahashi, T. Hokada, T. Kogiso, Y. Nishihara and T. Tsunogae. We also obtained invaluable comments from J. Ganguly, T. Kawakami and K. Sajeev on an earlier revision of the manuscript. Constructive criticisms from two reviewers (M. Arima and Y. Motoyoshi) and the editor (M. SatishKumar) are much appreciated. We thank V. J. Rajesh and P. Bhalla for help in improving the English of this paper. The experimental data for the Fe2þ –Mg partitioning between orthopyroxene and spinel used in this paper were obtained during the PhD work by K. S. at the Graduate School of Science and Engineering, Ehime University. The research was financially supported by the Sasakawa Scientific Research Grant from The Japan Science Society to K. S. (no. 15-144) and the Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of the Japanese Government to T. K. (14654093). This work was completed with the support of the 21st Century COE Program ‘How to build habitable planets’, Tokyo Institute of Technology, sponsored by the MEXT, Japan.
References B ABA , S. 2003. Two stages of sapphirine formation during prograde and retrograde metamorphism in the Palaeoproterozoic Lewisian Complex in South Harris, NW Scotland. Journal of Petrology, 44, 329–354. B ERMAN , R. G. & A RANOVICH , L. Y. 1996. Optimized standard state and solution properties of minerals ?. Model calibration for olivine, orthopyroxene, cordierite, garnet and ilmenite in the system
Fe2þ –Mg PARTITIONING BETWEEN Opx AND Spl FeO– MgO–CaO– Al2O3 – TiO2 –SiO2. Contributions to Mineralogy and Petrology, 126, 1 –24. B ERTRAND , P., E LLIS , D. J. & G REEN , D. H. 1991. The stability of sapphirine–quartz and hypersthene– sillimanite– quartz assemblages: an experimental investigation in the system FeO– MgO–Al2O3 –SiO2 under H2O and CO2 condition. Contributions to Mineralogy and Petrology, 108, 55–71. B OHLEN , S. R. & B OETTCHER , A. L. 1982. The quartz– coesite transformation: a precise determination and the effects of other components. Journal of Geophysical Research, 87, 7073– 7078. B OSE , S., F UKUOKA , M., S ENGUPTA , P. & D ASGUPTA , S. 2000. Evolution of high-Mg– Al granulite from Sunkarametta, Eastern Ghats, India: evidence for a lower crustal heating– cooling trajectory. Journal of Metamorphic Geology, 18, 223– 240. B RANDT , S., K LEMD , R. & O KRUSCH , M. 2003. Ultrahigh-temperature metamorphism and multistage evolution of garnet– orthopyroxene granulites from the Proterozoic Epupa Complex, NW Namibia. Journal of Petrology, 44, 1121– 1144. D ALLWITZ , W. B. 1968. Coexisting sapphirine and quartz in granulites from Enderby Land, Antarctica. Nature, 219, 476– 477. D AS , K., F UJINO , K., T OMIOKA , N. & M IURA , H. 2006. Experimental data on Fe and Mg partitioning between coexisting sapphirine and spinel: an empirical geothermometer and its application. European Journal of Mineralogy, 18, 49– 58. D EMING , W. E. 1943. Statistical Adjustment of Data. Wiley, New York. E LLIS , D. J. 1980. Osumilite– sapphirine –quartz granulites from Enderby Land, Antarctic: P– T conditions of metamorphism, implications for garnet–cordierite equilibria and the evolution of the deep crust. Contributions to Mineralogy and Petrology, 74, 201–210. E LLIS , D. J. 1987. Origin and evolution of granulites in normal and thickened crust. Geology, 15, 167–170. E LLIS , D. J., S HERATON , J. W., E NGLAND , R. N. & D ALLWITZ , W. B. 1980. Osumilite– sapphirine– quartz granulites from Enderby Land, Antarctica— mineral assemblage and reactions. Contributions to Mineralogy and Petrology, 72, 123–143. G ANGULY , J. & T IRONE , M. 1999. Diffusion closure temperature and age of a mineral with arbitrary extent of diffusion: theoretical formulation and applications. Earth and Planetary Science Letters, 170, 131– 140. G ANGULY , J., C HENG , W. & T IRONE , M. 1996. Thermodynamics of aluminosilicate garnet solid solution: new experimental data, an optimized model, and thermodynamic applications. Contributions to Mineralogy and Petrology, 126, 137–151. G ONCALVES , P., N ICOLLET , C. & M ONTEL , J. M. 2004. Petrology and in situ U–Th– Pb monazite geochronology of ultrahigh-temperature metamorphism from the Andriamena Mafic Unit, North–Central Madagascar. Significance of a petrographical P –T path in a polymetamorphic context. Journal of Petrology, 45, 1923–1957. G REW , E. S. 1980. Sapphirine þ uartz association from Archaean rocks in Enderby Land, Antarctica. American Mineralogist, 65, 821–836.
445
H ALL , H. T. 1971. Fixed points near room temperature. In: L LOYD , E. C. (ed.) Proceedings of Symposium of Accurate Characterization of High Pressure Environment. NBS Special Publication, 326, 313– 314. H ARLEY , S. L. 1984. An experimental study of the partitioning of the Fe and Mg between garnet and orthopyroxene. Contributions to Mineralogy and Petrology, 86, 359–373. H ARLEY , S. L. 1998a. On the occurrence and characterization of ultrahigh-temperature crustal metamorphism. In: T RELOAR , P. J. & O’B RIEN , P. J. (eds) What Drives Metamorphism and Metamorphic Reaction? Geological Society, London, Special Publications, 138, 81– 107. H ARLEY , S. L. 1998b. An appraisal of peak temperatures and thermal histories in ultrahigh-temperature (UHT) crustal metamorphism: The significance of aluminous orthopyroxene. Memoirs of National Institute of Polar Research, Special Issue, 53, 49–73. H ARLEY , S. L. & G REEN , D. H. 1982. Garnet– orthopyroxene barometry for granulites and peridotites. Nature, 300, 697– 701. H ARLEY , S. L. & M OTOYOSHI , Y. 2000. Al zoning in orthopyroxene in a sapphirine quartzite: evidence for .1120 8C UHT metamorphism in the Napier Complex, Antarctica, and implications for the entropy of sapphirine. Contributions to Mineralogy and Petrology, 138, 293– 307. H ARLEY , S. L., H ENSEN , B. J. & S HERATON , J. W. 1990. Two-stage decompression in Antarctica: orthopyroxene– sillimanite granulites from Forefinger Point, Enderby Land, implications for the evolution of the Archaean Napier Complex. Journal of Metamorphic Geology, 8, 591– 613. H ENSEN , B. J. & G REEN , D. H. 1971. Experimental study of the stability of cordierite and garnet in pelitic compositions at high pressures and temperatures. I: Compositions with excess alumino-silicate. Contributions to Mineralogy and Petrology, 33, 309– 330. H ENSEN , B. J. & G REEN , D. H. 1972. Experimental study of the stability of cordierite and garnet in pelitic compositions at high pressures and temperatures. II: Compositions without excess alumino-silicate. Contributions to Mineralogy and Petrology, 35, 331– 354. H ENSEN , B. J. & G REEN , D. H. 1973. Experimental study of the stability of cordierite and garnet in pelitic compositions at high pressure and temperatures. III. Synthesis of experimental data and geological applications. Contributions to Mineralogy and Petrology, 38, 151– 166. H OKADA , T., O SANAI , Y., T OYOSHIMA , T., O WADA , M., T SUNOGAE , T. & C ROWE , W. A. 1999. Petrology and metamorphism of sapphirine-bearing aluminous gneisses from Tonagh Island in the Napier Complex, East Antarctica. Polar Geoscience, 12, 49– 70. K AWASAKI , T. & M ATSUI , Y. 1983. Thermodynamic analyses of equilibria involving olivine, orthopyroxene and garnet. Geochimica et Cosmochimica Acta, 47, 1661– 1679. K AWASAKI , T. & M OTOYOSHI , Y. 2000. High-pressure and high-temperature phase relations of an orthopyroxene granulite from McIntyre Island, Enderby Land, East Antarctica. Polar Geoscience, 13, 114–134.
446
K. SATO ET AL.
K AWASAKI , T. & M OTOYOSHI , Y. 2006. Experimental constraints on the decompressional P –T paths of Rundva˚gshetta granulites, Lu¨tzow-Holm Complex, East Antarctica. In: F U¨ TTERER , D. K., D AMASKE , D., K LEINSCHMIDT , G., M ILLER , H. & T ESSENSOHN , F. (eds) Antarctica: Contributions to global earth sciences, Springer, 23– 36. K AWASAKI , T. & S ATO , K. 2002. Experimental study of Fe–Mg exchange reaction between orthopyroxene and sapphirine and its calibration as a geothermometer. Gondwana Research, 5, 741– 747. K AWASAKI , T., I SHIKAWA , M. & M OTOYOSHI , Y. 1993. A preliminary report on cordierite-bearing assemblages from Rundva˚gshetta, Lu¨tzow-Holm Bay, East Antarctica: Evidence for a decompressional P– T path? Proceedings of NIPR Symposium on Antarctic Geosciences, 6, 47– 56. K AWASAKI , T., S ATO , K. & M OTOYOSHI , Y. 2002. Experimental constraints on the thermal peak of a granulite from McIntyre Island, Enderby Land, East Antarctica. Gondwana Research, 5, 749–756. K OSYAKOVA , N. A., A RANOVICH , L. Y. & P ODLESSKII , K. K. 2004. Orthopyroxene– aluminous spinel equilibria in the system FeO–MgO– Al2O3 – SiO2: Experimental data and thermodynamic treatment. Lithos, 73, S63. K OSYAKOVA , N. A., A RANOVICH , L. Y. & P ODLESSKII , K. K. 2005. Equilibria of aluminous spinel with orthopyroxene in the system FeO–MgO– Al2O3 – SiO2: New experimental data and thermodynamic assessment. Doklady Earth Sciences, 400, 57–61. K RETZ , R. 1983. Symbols for rock-forming minerals. American Mineralogist, 68, 277–279. L EE , H. Y. & G ANGULY , J. 1988. Equilibrium composition of coexisting garnet and orthopyroxene: experimental determination in the system FeO– MgO– Al2O3 –SiO2, and applications. Journal of Petrology, 29, 93– 113. L IERMANN , H. P. & G ANGULY , J. 2003. Fe2þ – Mg fractionation between orthopyroxene and spinel: Experimental calibration in the system FeO– MgO– Al2O3 –Cr2O3 –SiO2, and applications. Contributions to Mineralogy and Petrology, 145, 217–227. M ATSUI , Y., S YONO , Y., A KIMOTO , S. & K ITAYAMA , K. 1968. Unit cell dimensions of some synthetic orthopyroxene group solid solutions. Geochemical Journal, 2, 61– 70. M IYAMOTO , T., Y OSHIMURA , Y., S ATO , K., M OTOYOSHI , Y., D UNKLEY , D. J. & C ARSON , C. J. 2004. Occurrences of metamorphosed ultramafic rock and associating rocks in Howard Hills, Enderby Land, East Antarctica: Evidence of partial melting from geochemical and isotopic characteristics. Polar Geoscience, 17, 88– 111. M OTOYOSHI , Y. & H ENSEN , B. J. 2001. F-rich phlogopite stability in ultra-high-temperature metapelites from the Napier Complex, East Antarctica. American Mineralogist, 86, 1404–1413. M OTOYOSHI , Y. & I SHIKAWA , M. 1997. Metamorphic and structural evolution of granulites from Rundva˚gshetta, Lu¨tzow-Holm Bay, East Antarctica. In: R ICCI , C. (ed.) The Antarctic Region: Geological Evolution and Processes. Terra Antartica, Siena, 65–72.
O KAY , A. I. 1994. Sapphirine and Ti-clinohumite in ultra-high-pressure garnet-pyroxenite and eclogite from Dabie Shan, China. Contributions to Mineralogy and Petrology, 116, 145–155. O’N EILL , H. St. C. , 1986. No-MoO2 (MOM) oxygen buffer and the free energy of formation of MoO2. American Mineralogist, 71, 1007– 1010. O SANAI , Y., H AMAMOTO , T., M AISHIMA , O. & K AGAMI , H. 1998. Sapphirine-bearing granulites and related high-temperature metamorphic rocks from the Higo metamorphic terrane, west–central Kyushu, Japan. Journal of Metamorphic Geology, 16, 53–66. O UZEGANE , K., G UIRAUD , M. & K IENAST , J. R. 2003. Prograde and retrograde evolution in high-temperature corundum granulites (FMAS and KFMASH systems) from In Ouzzal Terrane (NW Hoggar, Algeria). Journal of Petrology, 44, 517–545. R OBIE , R. A., H EMINGWAY , B. S. & F ISHER , J. R. 1978. Thermodynamic properties of minerals and related substances at 298.15 K and 1 Bar (10 5 Pascals) pressure and at higher temperatures. US Geological Survey Bulletin, 1452. S AJEEV , K. & O SANAI , Y. 2004. Ultrahigh-temperature metamorphism (1150 8C, 12 kbar) and multistage evolution of Mg– Al– rich granulites from the Central Highland Complex, Sri Lanka. Journal of Petrology, 45, 1821–1844. S AJEEV , K., O SANAI , Y. & S ANTOSH , M. 2004. Ultrahigh-temperature metamorphism followed by two-stage decompression of garnet–orthopyroxene – sillimanite granulites from Ganguvarpatti, Madurai block, southern India. Contributions to Mineralogy and Petrology, 148, 29–46. S AKAI , S. & K AWASAKI , T. 1997. An experimental study of Fe–Mg partitionings between orthopyroxene and cordierite in the Mg-rich portion of the Mg3Al2Si3O12 – Fe3Al2Si3O12 system at atmospheric pressure: calibration of its geothermometry for hightemperature granulites and igneous rocks. Proceedings of NIPR Symposium on Antarctic Geosciences, 10, 165–177. S ATO , K. & K AWASAKI , T. 2002. High-pressure and high-temperature experiments on the phase relations in the system of Mg-rich garnet composition (Prp75Alm25): Implication for the Fe–Mg partitioning between garnet and orthopyroxene. Polar Geoscience, 15, 66–79. S ATO , K., M IYAMOTO , T. & K AWASAKI , T. 2004. Experimental constraints of metamorphic pressure and temperature, and phase relations of a phlogopite-bearing orthopyroxene granulite from Howard Hills, Napier Complex, East Antarctica. Journal of Mineralogical and Petrological Sciences, 99, 191–201. S ATO , K., M IYAMOTO , T. & K AWASAKI , T. 2006. Experimental calibration of sapphirine– spinel Fe2þ – Mg exchange thermometer: Implication for constraints on P– T condition of Howard Hills, Napier Complex, East Antarctica. Gondwana Research, 9, 398– 408. S ENGUPTA , P., S EN , J., D ASGUPTA , S., R AITH , M., B HUI , U. K. & E HL , J. 1999. Ultra-high temperature metamorphism of metapelitic granulites from Kondapalle, Eastern Ghats Belt: Implications for the
Fe2þ –Mg PARTITIONING BETWEEN Opx AND Spl Indo-Antarctic correlation. Journal of Petrology, 40, 1065–1087. W OOD , B. J. & N ICHOLLS , J. 1978. The thermodynamic properties of reciprocal solid solutions. Contributions to Mineralogy and Petrology, 66, 389–400.
447
Y OSHIMURA , Y., M OTOYOSHI , Y., G REW , E. S., M IYAMOTO , T., C ARSON , C. J. & D UNKLEY , D. J. 2000. Ultrahigh-temperature metamorphic rocks from Howard Hills in the Napier Complex, East Antarctica. Polar Geoscience, 13, 60–85.
Index Note: Page numbers in italics refer to Figures. Page numbers in bold refer to Tables. adakite see meta-tonalite, adakitic Akai-Misaki, geochronology study 167 Akarui Point 148, 212, 213, 215 deformation history 354 geochronology study 168 kornerupine occurrence 352 boron sources 367, 369–371 chemistry methods of analysis 354– 355 results 358– 359, 361 description 355–357, 360 P –T evolution 364 petrological setting 355 replacement reactions 364 –367 Akebono Rock 212 Albany–Fraser Range 5, 166, 178 Alexander von Humboldt Gebirge 71 allanite, Tonagh Island 295, 305 alteration zones see metasomatism and alteration amphibole see grunerite also hornblende amphibolite facies Lu¨tzow-Holm Complex 211, 212, 352, 377, 378 Nampula Complex 72 Namuno Block 95 Napier Complex 284 Sør Rondane Mountains 24, 60 Amphitheatre Lakes, Nd model ages 56 Amundsen Bay 139, 258 geological setting 317, 319– 321 see also fluid inclusion studies, Tonagh Island Amundsen Bay Fault 144, 145 Amundsen Dykes 122– 123, 139, 196, 258, 259 Angonia Complex 93, 95 Angonia Group 95 Annandagstoppane 71 anorthosite magmatism 71 Antarctic Plate 1 apatite, Tonagh Island 295, 304 40 Ar/39Ar Cape Hinode meta-tonalite 338 Lu¨tzow-Holm Complex chronology methods of analysis 173 results 169, 174, 176 Armlenet 80– 81 Arrhenius Equation 420 Austhovde 148, 212 geochronology study 168, 169 backscatter electron (BSE) imaging 26 Balchen, Nd model ages 53 Beaver Glacier Fault 144, 145 Beaver Island 123, 129 Belgica Mountains 53 Bergin, Mount 123 Berrodden 169 biotite chemistry at Akarui Point methods of analysis 355 results 362– 363
gneiss of Rundva˚gshetta mineral chemistry 381, 383, 384 petrography 378, 380 significance in metamorphic history 385– 389 Tonagh Island 296, 305 xenocrysts in Cape Hinode meta-tonalite 338–345 bleach zones see metasomatism and alteration Borgmassivet 71 borosilicate see Akarui Point Botnnuten, geochronology study 169 Bowl Island 123 Brattnipene 53 Breidvagnipa 148 geochronology study 167 marble study 150, 153 Bunger Hills 5, 166, 178, 195 Bunt Island 123 geological setting 319– 320 metamorphic age 129 see also fluid inclusion studies, geothermometry Byobu Rock 148 d13C marbles from Lu¨tzow-Holm Complex methods of analysis 152 results 153, 155– 157 carbonates see marbles carbonic fluids see fluid inclusion studies Casey Bay 254 cathodoluminescence (CL) imaging 26, 75, 150–151, 152 charnockite 72, 81–83, 84– 85, 177 CHIME (chemical U– Th–Pb isochron method) Lu¨tzow-Holm Complex 173 Napier Complex 125–126, 128–129 Chiure Supergroup 95 Christmas Point 127, 128 Circum-Antarctic Mobile Belt 3 Circum-East Antarctic Belt 22 clinopyroxene central Dronning Maud Land chemistry 406, 407, 408, 409 Tonagh Island 294, 304 xenocrysts in Cape Hinode meta-tonalite 338 –345 clinozoisite, Tonagh Island 295, 305 Condon Hills 56, 123 cordierite granulite of Rundva˚gshetta mineral chemistry 381, 383, 385 significance in metamorphic history 385– 389 cristobalite 423 Cronus, Mount 123 crystal size studies see under Skallen Culicui Suite 95– 96, 96, 101 Dallwitz Nunatak 123 dolerite see dykes
450 Dronning Maud Land correlations with Mozambique geochronology 101, 102, 104, 107 metamorphism 108, 110–111 structures 105, 106–108, 111 summary 111–115 East African–Antarctic Orogen 69 granite geochronology methods of analysis 75 results 80–83 historical data 73 SIMS 76– 79 results discussed tectonic interpretation 84– 86 tectonic model 86 summary 87 Late Neoproterozoic– Early Palaeozoic orogenesis 71–72 Dronning Maud Land (central) 5 metamorphic history compared with Sør Rondane Mountains 62–63 Nd model ages 57 SHRIMP ages 60 see also Mu¨hlig– Hofmannfjella Mountains; Schirmacher Hills Dronning Maud Land (central–eastern) geological setting 236–239 see also dykes Dronning Maud Land (eastern) see Lu¨tzow-Holm Complex; Sør Rondane Mountains Dronning Maud Land (western), metamorphic history compared with Sør Rondane Mountains 62– 63 Dufek 53 dykes Cape Hinode 335 central–eastern Dronning Maud Land geochemistry 240–242, 243, 244 isotope analysis methods 243, 245 results 245–246, 247 petrography 240–242 summary crust/mantle contribution 246– 247 residual phases 247– 248 tectonic setting 248– 250 Riiser-Larsen, Mount Amundsen 122–123, 196, 258, 259 Proterozoic field setting 197 –198 geochemistry methods of analysis 199, 201 results 200, 201–204, 204– 206 results discussed 206– 207 summary 207 petrography 198 East African Orogeny 23 East African– Antarctic Orogen (EAAO) 69, 401 Dronning Maud Land 71–72 modelling orogen behaviour 86– 87 Mozambique 72 East Antarctic Shield 3
INDEX East Gondwana, amalgamation of 1, 3, 9, 23 East Ongul 54, 55, 167, 168, 169 Eastern Ghats Belt 5, 166, 178 Edward VIII Gulf 258 Edwards Island 123 elastic property measurement methods 185 results 185–190 results discussed 190 –192 specimens 184 summary 192–193 Enderby Land 184 see also Napier Complex epidote 305 fayalite 407 feldspar see orthoclase also plagioclase Field Island 123 Filchnerfjella 71 map 238, 403 metamorphic P– T path 410, 412 metamorphic textures 413 mineral chemistry 407, 409 petrography 405 –407 regional correlation of P– T path 413– 416 thermobarometry 411 Fletta, Mount 56 Fleynoya 55 fluid inclusion studies 318 methods 321–323 results 323–326 results discussed 326 –329 Forefinger Point 123 metamorphic temperatures 443 form-line mapping interpretation 142 method 140–142 results discussed 142 –145 Fyfe Hills 57, 123, 254, 255 Gage Ridge 123, 254, 255 garnet central Dronning Maud Land chemistry 405, 406, 407, 408, 409 gneiss of Rundva˚gshetta mineral chemistry 381– 382, 385 petrography 378, 380–381 significance in metamorphic history 385–389 Tonagh Island 294, 304 xenocrysts in Cape Hinode meta-tonalite 338–345 Gawler Craton 3 Geci Group 111 geochemistry, Cape Hinde meta-tonalite 336–337 Geoffrey Hills 123 geochronology see 40Ar/39Ar; K–Ar; Rb–Sr; Sm–Nd; U–Pb geothermometry Fe–Mg partitioning introduction 431–432 method 432– 433 results 433–436 results discussed 441– 444 thermodynamic model 437–441
INDEX TiO2 in quartz method 421, 423 results 422, 423–426 results discussed 428–429 theory 420– 421 see also microthermometry Gjelsvikfjella 71, 413, 414 map 238 gneiss see Skallen gneiss Rundva˚gshetta, petrography 378–381 see also elastic property measurement, Skallen petrology Gobanme Rock 212 Gondwana, assembly 147, 177, 211, 401 granite and granitoid magmatism 402 EAAO of Dronning Maud Land and Mozambique compared geological setting 71– 75 methods of analysis 75 results 80– 84 historical data 73 SIMS 76–79 results discussed tectonic interpretation 84–86 tectonic model 86 summary 87 granulite see elastic property measurement, fluid inclusion studies, geothermometry, granulite facies Lu¨tzow-Holm Complex 211, 212 Nampula Complex 72, 377, 378 Namuno Block 95 Sør Rondane Mountains 24, 60 Schirmacher Nappe 71 Grenville event 22, 23 Grunehogna 5 Grunehogna Craton 71 grunerite, Tonagh Island 296 haematite, xenocrysts in Cape Hinode meta-tonalite 341 Heimfrontfjella 71 Henry’s Law 420 Highland Complex 105, 110, 111, 177 Hinode, Cape 148, 212 geochemistry 336– 337 geochronology 335, 338 40 Ar/39Ar 338 K– Ar 338 Rb– Sr 335 SHRIMP U– Pb zircon 335 Sm–Nd 335 geochronology study 167, 168, 169 geological setting 334 meta-tonalite origins 346–347 meta-tonalite tectonic significance 347 Nd model ages 54 palaeomagnetism 338 petrology calc-silicates 334 dyke rocks 335 gneisses 335
meta-tonalite 334 xenocryst studies 338– 345 Hochlinfjellet 413, 415 hornblende central Dronning Maud Land chemistry 405, 406, 407, 408, 409 Tonagh Island 296, 305 xenocrysts in Cape Hinode meta-tonalite 338 –345 Howard Hills 123, 127 metamorphic age 129 metamorphic temperatures 443 humite 150, 152 Hydrographer Island 123 ICP-MS Lu¨tzow-Holm Complex 152 Dronning Maud Land 245 Mozambique geochronology 97, 98– 100 Napier Complex 199, 200 igneous activity 166, 178 see also dykes; granite; metabasite; pegmatites ilmenite 341 Innhovde 55, 168, 169 intrusives 166, 178 see also dykes; granite; metabasite; pegmatites iron –magnesium (Fe–Mg) exchange thermometry see under geothermometry isotopes, stable see d13C; d18O isotopic dating see 40Ar/39Ar; K –Ar; Rb– Sr; Sm– Nd; U– Pb JARE schedule 122 Jutulsessen 402 map 403 metamorphic P –T path 410, 412 metamorphic textures 413 mineral chemistry 407, 408, 409 petrography 404–405 regional correlation of P– T path 413– 416 thermobarometry 411 K –Ar Cape Hinode meta-tonalite 338 Lu¨tzow-Holm Complex chronology methods 173 results 168, 169, 174, 176 Kabuto Rock 148, 150 Kasumi Rock 148 geochronology study 167, 168, 169 marble study 150, 152, 153 Nd model ages 54 Kemp Coast 258 Khmara dykes 196 Kirwanveggen 105, 106, 108, 110 kornerupine see Akarui Point Kuunga Orogeny 23 kyanite 211, 338– 345 Lalamo Complex 72, 95 lamproite see dykes lamprophyre see dykes
451
452
INDEX
Langhovde 212 Lars Christensen Expedition (1937) 21 Larsemann Hills 352 Layered Gneiss Series 196 Leewin Complex 5 lineations, Dronning Maud Land 111 Lira, Mount 123 Lu–Hf 129 Lurio Belt 72, 92–93, 95, 103 –105 Lurio Supergroup 95 Lu¨tzow-Holm Complex 5, 166, 190, 192, 333, 378, 391 geological setting 148–149, 211–212 metamorphic history compared with Sør Rondane Mountains 61–62 Nd model ages 53, 54, 55 SHRIMP ages 60 UHT metamorphism 22 see also Akarui Point; Hinode, Cape; marbles; metabasites; Rundva˚gshetta; Skallen McIntyre Island 123 geothermometry 419, 422, 424, 425, 426, 427, 428 MacRobertson Land 178, 195 magmatism see dykes; granite; metabasites; pegmatites magnesium (Mg) see geothermometry magnetic anomaly map 145 magnetite, xenocrysts in Cape Hinode meta-tonalite 341 Main Shear Zone 238, 239 Malawi, correlation with Antarctica 97, 98, 99, 100 Mamala Gneiss 96 marbles 149–150 field relations 150 geochemical analysis methods 150– 152 results carbonate mineral chemistry 154 carbonate trace elements 154– 155 isotope chemistry 153, 155–157 mineralogy and texture 152– 154 results discussed depositional age 159–161 post-depositional alteration 158–159 sedimentary features 159 Marrupa Complex 72, 95 Massive Gneiss Series 196 Maud Belt 3 Mawson Continent 3 Mejell 53 Meluco Complex 95 meta-tonalite, adakitic geochemistry 336– 337 petrology 334 xenocryst studies 338– 345 discussion of results 346– 347 metabasite definition 213 field occurrence 213 geochemistry methods of analysis 215, 217 results 218, 219 alteration 217 isotope composition 220, 224–225
major and trace elements 217, 221, 222 –224 results discussed magmatic process 227 Nd constraints 228– 230 protoliths 225– 226 tectonic setting 226–227 summary 230 petrography 214 –215, 216 metasomatism and alteration 230 analytical methods 290, 292, 294 results bulk rock geochemistry, 297, 298, 299, 300, 301, 302, 303 host rock mineralogy 291 mineral chemistry 293, 294, 295, 296, 303–305 mineral modal analysis 288 monazite geochronology 297, 305 P –T estimations 294, 295, 296 photomicrographs 289 whole rock reactions 305– 308 results discussed alkali and alkaline earth element reactions 308–310 rare earth element and Y patterns 310– 311 tectonic significance 311–312 sampling pattern 287, 290 summary 313 mica see biotite also muscovite microthermometry fluid inclusion studies method 321– 323 results 323–326 results discussed 326 –329 Miller Range 5 mineral assemblage mapping, Napier Complex 144 Mizuho Plateau 190– 191 Mocuba klippen 72 Mocuba Suite 95, 96, 101, 109 Molucue Group 96 Monapo Complex 93, 96, 97, 105, 106, 109 Monapo klippen 72 monazite geochronology, Tonagh Island 297, 305 Montepuez Complex 95 Mozambique correlations with Antarctica and Sri Lanka 111– 115 crustal structure 93–95 East African– Antarctic Orogen 69, 72 geochronology 96– 101, 104 granite geochronology methods of analysis 75 results 83– 84 historical data 73 SIMS 76–79 results discussed tectonic interpretation 84–86 tectonic model 86 summary 87 metamorphic history 108 –110 national geological mapping 91– 92 rock types 95– 96 status of Lurio Belt 92– 93, 95, 106 structures 103, 105, 106, 109
INDEX Mozambique Belt 5 Mozambique Ocean 161, 212, 236 Mozambique Suture 236 Mramornye nunatak 105, 106, 107 M’Sawize Complex 72, 95 Msawize Group 95 Muaquia Complex 95 Mugeba Complex 93, 96, 97, 105, 106, 109 Mu¨hlig-Hofmannfjella Mountains 71, 101, 105, 107, 108 geological setting 236–237 mafic dyke study geochemistry 240, 241, 243, 244 isotope analysis 245 –246, 247 petrography 240–242 summary crust/mantle contribution 246 –247 residual phases 247–248 tectonic setting 248 –250 map 238 Murrupula Suite 96 muscovite 296 Nairoto Complex 95 Namama Shear Zone 106 Namaqua–Natal Belt 5 Nampula Block 93, 95–96, 100, 102, 108– 109, 111 Nampula Complex 72 Namuno Block 93, 95, 96–97, 102 Napier Complex 5, 166, 178 Napier Complex see Bunt Island Napier Complex form-line mapping interpretation 142 method 140– 142 results discussed 142–145 geological setting 139–140, 253–255, 258, 284 –286 magnetic anomalies 145 mineral assessment map 144 Nd model ages 56, 57 protolith age 121, 125–126 role of TTG 130– 131 summary history 132– 133, 255 UHT granulite see geothermometry UHT metamorphism 121 age 126–127, 128– 129, 130 timing 131–132 see also Amundsen Bay; Bunt Island; Priestley Peak; Riisier Larsen, Mount; Tonagh Island Napier Fault 144, 145 Napier Mountains 123, 254, 255 Nd isotopes, see Sm–Nd Nesoya geochronology study 167 Nd model ages 54 Niban Rock 213 Novolazarevskaya Station 71 nunataks, plutonic 74–75 d18O marbles from Lu¨tzow-Holm Complex methods of analysis 152 results 155– 157
453
Ocua Complex 95 Oddesteinen 81 Oku-Iwa Rock 212 geochronology study 167 Nd model ages 54 Oldfield, Mount 123 Omega, Cape 212 geochronology study 167 Ongul Island 183, 212 Ongul Strait 55 Operation Highjump (1946–7) 21 orthoclase 293, 303 –304 orthopyroxene central Dronning Maud Land chemistry 405, 406, 407, 408, 409 see geothermometry granulite of Rundva˚gshetta mineral chemistry 381, 382, 385 significance in metamorphic history 385– 389 Tonagh Island 294, 304 xenocrysts in Cape Hinode meta-tonalite 338–345 Orvinfjella 71, 72 granitoid intrusion 402 map 238 metamorphic facies 401– 402 metamorphic P –T path 414 Nd model ages 57 Orvinfjella shear zone 105 Otto von Gruber Gebirge 71 Oygarden Islands 258, 352 P waves see elastic property measurement palaeomagnetism, Cape Hinode 338 Pan African Orogeny 22, 23, 147, 177, 190, 211, 237, 401 Pardoe, Mount 123 metamorphic age 128 partitioning experiments see geothermometry Pb/Pb evaporation 97, 99, 100 pegmatites and metasomatic alteration 285, 287 see metasomatism and alteration Pingvinane 53 plagioclase central Dronning Maud Land chemistry 405, 406, 407, 408, 409 chemistry at Akarui Point methods of analysis 355 results 364 granulite of Rundva˚gshetta mineral chemistry 381, 384, 385 significance in metamorphic history 385– 389 Tonagh Island 293, 303– 304 xenocrysts in Cape Hinode meta-tonalite 338–345 Priestley Peak 123, 196 fluid inclusion studies 318 method 321–323 results 324, 326 results discussed 326 –329 geological setting 321 Prince Charles Mountains 178, 195 Prince Olav Coast see Lu¨tzow-Holm Complex Princess Astrid Coast see Schirmacher Hills Princess Elizabeth Land 195
454 Proclamation Island 254 Proto-Kalahari Craton 71 Prydz Bay 3, 5 pyroxene see clinopyroxene also orthopyroxene pyroxenite see elastic property measurement quartz, rutilated geothermometer method 421, 423 results 422, 423–426 results discussed 428–429 theory 420–421 Queen Mary Land 195 Raman spectroscopy, in fluid inclusion studies 321– 323 Raoult’s Law 420 Rapale Gneiss 96 rare earth element (REE) geochemistry 203 Rayner Complex 3, 5, 121, 122, 139, 166, 178, 190, 192 form-line mapping interpretation 142 method 140–142 results discussed 142–145 metamorphic history compared with Sør Rondane Mountains 61–62 metamorphic temperatures 443 Nd model ages 55, 56 SHRIMP ages 60 Rayner Province 5, 166, 178 Rb–Sr Cape Hinode meta-tonalite 338 Dronning Maud Land 243, 245 Lu¨tzow-Holm Complex chronology methods of analysis 173 results 167– 169 results discussed 175, 176 Lu¨tzow-Holm Complex metabasite methods of analysis 217 results 220, 224–225 Napier Complex 199, 129, 204– 206 Riiser-Larsen dykes 195, 204 –206 Reference Peak 129 Riiser-Larsen Main Shear Zone 124, 144, 145, 260 timing of 274 –275, 278 Riiser-Larsen, Mount 123 basement 196– 197 geological setting 255, 258– 260 geothermometry 419, 422, 424, 425, 427, 428 map 197, 256 metamorphic age 128, 129 Nd model ages 56 protolith 125 structures 124, 259 UHT metamorphism 123–124 see also dykes, sapphirine, elastic property measurement, Rodinia 3, 71 Ruker Terrane 5 Rundva˚gshetta 148, 212
INDEX geochronology study history of research 166, 170 methods of analysis 173 results 168, 169, 174 results discussed 174– 178 sample description 171– 173 summary 178– 179 geological setting 378 geothermometry 419, 422, 424, 425, 427, 428 metamorphic temperatures 443 mineral chemistry methods of analysis 381 results biotite 383, 384 cordierite 383, 385 garnet 381–382 orthopyroxene 382 plagioclase 384, 385 sapphirine 382– 383 Nd model ages 55 petrographic studies garnet–biotite gneiss 378, 380 garnet–sillimanite gneiss 380– 381 sapphirine granulite 378 significance of metamorphic history 385 –389 significance of thermal maximum for UHT metamorphism 165, 166, 211, 213, 354 Rundvagskollane 212 rutile xenocrysts in Cape Hinode meta-tonalite 342 see also geothermometry S waves see elastic property measurement Sanagoe thrust zone 95, 105 Sanbagawa Belt, geothermometry 419, 422, 424, 425, 426 Sandercock Nunataks 56 sapphirine chemistry at Akarui Point methods of analysis 355 results 365 granulite of Rundva˚gshetta mineral chemistry 381, 382–383 petrography 378 significance in metamorphic history 385–389 Napier Complex 139 Riiser-Larsen, Mount geochemistry 264, 265, 266– 267 mineral associations 261 P –T constraints 268–272 petrography 260, 262–263, 266 significance for metamorphic gradient 278–279 significance for UHT metamorphism 273 –274 textures 266 Schirmacher Hills 101, 104, 105, 106, 108, 110, 111 map 403 metamorphic facies 402 metamorphic P– T conditions 407, 410, 412 metamorphic textures 410, 413 mineral chemistry 407, 408 Nd model ages 57 petrography 404 regional correlation of P– T path 413– 416
INDEX sapphirine occurrence 402 SHRIMP ages 60 Sm– Nd age 402 thermobarometry 411 Schirmacher Nappe 71 Schirmacher Oasis 235, 237 Schneide 81, 83 Scotia Plate 1 Scott Mountains Fault 144 SEAL (Structure and Evolution of the East Antarctic Lithosphere) project 122, 183 seismic waves see elastic property measurement Shackleton Range 5 shield concept 3 SHRIMP (sensitive high-resolution ion microprobe) Cape Hinode meta-tonalite 335 Dronning Maud Land 105, 106 Lu¨tzow-Holm Complex chronology 354, 378 methods 173 results 168 Mozambique geochronology 97, 98–100, 102 Mu¨hlig-Hofmannfjella Mountains 107, 108 Napier Complex 125– 126, 128 –129 Sør Rondane Mountains methods 26 results 27– 33, 31–39, 40–42 results discussed 33, 39, 43–50 summary 50–52 sillimanite gneiss of Rundva˚gshetta petrography 380–381 significance in metamorphic history 385–389 Lu¨tzow-Holm Complex 211 SIMS (secondary ionization mass spectrometry) Napier Complex 125– 126, 128 –129 Sinnan Rock geochronology study 167, 168 Nd model ages 53, 54 Skallen 148, 212 geochronology study history of research 166, 170 methods of analysis 173 results 167, 168, 174 results discussed 174–178 sample description 170– 171 summary 178– 179 gneiss petrology crystal size studies methods of analysis 392, 394–395 results 393, 394, 395 results discussed annealing effects 395– 397 nucleation and growth 397–398 field occurrence 391– 392 marble study methods 150– 152 results 152– 157 results discussed 157–161 Nd model ages 55 P– T fluid evolution 149 Skallevikshalsen 148, 212 geochronology study history of research 166, 170
455
methods of analysis 173 results 167, 174 results discussed 174 –178 sample description 171 summary 178–179 marble study 150 methods 150–152 results 153, 154 Nd model ages 55 Skarvsnes 148, 212 geochronology study 167 Nd model ages 55 Sm– Nd Dronning Maud Land mafic dykes methods of analysis 243, 245 results 245–246, 247 Lu¨tzow-Holm Complex chronology methods 173, 199 results 167, 168, 176, 177, 204–206, 228–230 Lu¨tzow-Holm Complex metabasite methods of analysis 217 results 220, 224– 225 Mozambique geochronology 97 Napier Complex methods of analysis 125, 199 results 128–129, 204–206 Namuno Block 99 Riiser-Larsen dykes 195, 204– 206 Sør Rondane Mountains methods 52 results 53–57 results discussed 52, 58– 59 Sones, Mount 123, 254, 255 Nd model ages 56, 57 Sør Rondane Mountains 5, 101, 104, 105, 110, 111 geological setting 237– 239 granitoids 402 igneous activity 166, 178 mafic dykes study geochemistry 241, 242, 243, 244 isotope analysis 245– 246, 247 petrography 240 –242 summary crust/mantle contribution 246–247 residual phases 247– 248 summary tectonic setting 248– 250 map 25 metamorphic history 24 Nd model ages method 52 results 53– 57 results discussed 52, 58–59 SHRIMP U–Pb chronology method 26 results 27– 32, 34–39, 40– 42 results discussed NE terrane 33, 39, 43– 48 SW terrane 48– 50 summary 50– 52 tectonothermal events 59– 61 terrane comparisons 61– 63 Sør Rondane Suture 25, 238, 239 Southern Irumide Belt 93, 97, 98, 99 Southern Irumide Complex 95
456
INDEX
spinel 366 see also geothermometry Sr isotopes see Rb-Sr Sri Lanka correlations with Antarctica geochronology 101–102, 104, 177 metamorphism 105, 110, 111, 113 Stillwell Hills 258 structure mapping see form-line mapping Sverdrupfjella 105, 106–107, 108, 110, 112 granitoids 402, 413, 414 Syowa Station 2, 148 Tange Promontory 123 Telen geochronology study 168 Nd model ages 55 Terre Adelie 5 Tete Complex 94, 105 Th–Pb, Tonagh Island 297, 305 Thala Hills 56 thermodynamic modelling Fe– Mg exchange thermometry 437 –441 TiO2 thermometer 420– 421 TIMS (thermal ionization mass spectrometry) 243 Dronning Maud Land 243 Lu¨tzow-Holm Complex 173, 217 Napier Complex 199 Namuno Block 98 Mozambique geochronology 97, 98–100 TiO2 in quartz geothermometer see under geothermometry Tippet Nunataks Fault 145 TitaniQ thermometer 421, 429 titanite see U–Pb– zircon– titanite Tod, Mount 123 Tonagh Island 123, 196, 254, 255 geological setting 319 history of research 286– 287 metamorphic age 128 metamorphic temperatures 443 metamorphism 124 Nd model ages 56 pegmatite occurrence 285, 287 protolith 125 structure 124 see also metasomatism and alteration, fluid inclusion studies tonalite–trondhjemite –granodiorite (TTG) 130 –131, 192 tourmaline 368 Trail, Mount 123 transform faults 1 Transitional Gneiss Series 196 Troll Station 402 mineral chemistry 407, 408, 409 petrography 404– 405 Tula Mountains Fault 144, 145 U–Pb– zircon– titanite age Cape Hinode meta-tonalite 335 Dronning Maud Land methods 75 results 76–79, 80– 84, 105, 106 results discussed 84– 86 summary 87
Lu¨tzow-Holm Complex 354, 378 methods of analysis 173 results 168 Mozambique geochronology 97, 98– 100, 102 Mu¨hlig-Hofmannfjella Mountains 107, 108 Napier Complex method 125– 126 results 128, 129 Sør Rondane Mountains methods 26 results 27– 33, 31–39, 40– 42 results discussed 33, 39, 43– 50 summary 50– 52 UHT (ultrahigh temperature) metamorphism experimental studies see microthermometry; geothermometry; sapphirine; localities Napier Complex 121, 184, 254, 284 Riiser-Larsen, Mount 123, 184, 255, 273–274 Tonagh Island 286– 287 see also Rundva˚gshetta as thermal maximum locality ultrapotassic igneous activity 166, 178 Unango Complex 72, 95 Underwood, Mount 55 uralitization 196 Urfjell Group 106 Utholmen 55 Vechernaya, Mount 55 velocity studies see elastic property measurement Vestfold Hills 5, 195 Vijayan Complex 104, 105 Ward Nunataks 56 Wegener– Mawson Mobile Belt 3 West Gondwana 1, 3, 9 West Ongul 55 Wilkes Province 3, 5, 166, 178 Windmill Island 5, 166, 178, 352 Wohlthaat Massif 71, 72, 101, 105 granitoid intrusion 402 metamorphic facies 401– 402 Nd model ages 57 xenocryst studies Cape Hinode meta-tonalite 338–345 significance of results 346 –347 Xixano Complex 72, 95 XRF 199, 200, 243 Yamato– Belgica Complex igneous activity 22, 166, 178 metamorphic history compared with Sør Rondane Mountains 61– 62 Nd model ages 53 SHRIMP ages 60 Yuzhnaya, Mount 123 Zambia, correlation with Antarctica 97 zircon dating see U–Pb– zircon–titanite Zircon Point 123
